Showing 1–50 of 180 results for author: Soljačić, M

OccamLLM: Fast and Exact Language Model Arithmetic in a Single Step

Authors: Owen Dugan, Donato Manuel Jimenez Beneto, Charlotte Loh, Zhuo Chen, Rumen Dangovski, Marin Soljačić

Abstract: Despite significant advancements in text generation and reasoning, Large Language Models (LLMs) still face challenges in accurately performing complex arithmetic operations. To achieve accurate calculations, language model systems often enable LLMs to generate code for arithmetic operations. However, this approach compromises speed and security and, if finetuning is involved, risks the language mo… ▽ More Despite significant advancements in text generation and reasoning, Large Language Models (LLMs) still face challenges in accurately performing complex arithmetic operations. To achieve accurate calculations, language model systems often enable LLMs to generate code for arithmetic operations. However, this approach compromises speed and security and, if finetuning is involved, risks the language model losing prior capabilities. We propose a framework that enables exact arithmetic in \textit{a single autoregressive step}, providing faster, more secure, and more interpretable LLM systems with arithmetic capabilities. We use the hidden states of an LLM to control a symbolic architecture which performs arithmetic. Our implementation using Llama 3 8B Instruct with OccamNet as a symbolic model (OccamLlama) achieves 100\% accuracy on single arithmetic operations ($+,-,\times,÷,\sin{},\cos{},\log{},\exp{},\sqrt{}$), outperforming GPT 4o and on par with GPT 4o using a code interpreter. OccamLlama also outperforms GPT 4o both with and without a code interpreter on mathematical problem solving benchmarks involving challenging arithmetic, thus enabling small LLMs to match the arithmetic performance of even much larger models. We will make our code public shortly. △ Less

Submitted 29 June, 2024; v1 submitted 4 June, 2024; originally announced June 2024.

Stochastic logic in biased coupled photonic probabilistic bits

Authors: Michael Horodynski, Charles Roques-Carmes, Yannick Salamin, Seou Choi, Jamison Sloan, Di Luo, Marin Soljačić

Abstract: Optical computing often employs tailor-made hardware to implement specific algorithms, trading generality for improved performance in key aspects like speed and power efficiency. An important computing approach that is still missing its corresponding optical hardware is probabilistic computing, used e.g. for solving difficult combinatorial optimization problems. In this study, we propose an experi… ▽ More Optical computing often employs tailor-made hardware to implement specific algorithms, trading generality for improved performance in key aspects like speed and power efficiency. An important computing approach that is still missing its corresponding optical hardware is probabilistic computing, used e.g. for solving difficult combinatorial optimization problems. In this study, we propose an experimentally viable photonic approach to solve arbitrary probabilistic computing problems. Our method relies on the insight that coherent Ising machines composed of coupled and biased optical parametric oscillators can emulate stochastic logic. We demonstrate the feasibility of our approach by using numerical simulations equivalent to the full density matrix formulation of coupled optical parametric oscillators. △ Less

Submitted 6 June, 2024; originally announced June 2024.

QuanTA: Efficient High-Rank Fine-Tuning of LLMs with Quantum-Informed Tensor Adaptation

Authors: Zhuo Chen, Rumen Dangovski, Charlotte Loh, Owen Dugan, Di Luo, Marin Soljačić

Abstract: We propose Quantum-informed Tensor Adaptation (QuanTA), a novel, easy-to-implement, fine-tuning method with no inference overhead for large-scale pre-trained language models. By leveraging quantum-inspired methods derived from quantum circuit structures, QuanTA enables efficient high-rank fine-tuning, surpassing the limitations of Low-Rank Adaptation (LoRA)--low-rank approximation may fail for com… ▽ More We propose Quantum-informed Tensor Adaptation (QuanTA), a novel, easy-to-implement, fine-tuning method with no inference overhead for large-scale pre-trained language models. By leveraging quantum-inspired methods derived from quantum circuit structures, QuanTA enables efficient high-rank fine-tuning, surpassing the limitations of Low-Rank Adaptation (LoRA)--low-rank approximation may fail for complicated downstream tasks. Our approach is theoretically supported by the universality theorem and the rank representation theorem to achieve efficient high-rank adaptations. Experiments demonstrate that QuanTA significantly enhances commonsense reasoning, arithmetic reasoning, and scalability compared to traditional methods. Furthermore, QuanTA shows superior performance with fewer trainable parameters compared to other approaches and can be designed to integrate with existing fine-tuning algorithms for further improvement, providing a scalable and efficient solution for fine-tuning large language models and advancing state-of-the-art in natural language processing. △ Less

Submitted 31 May, 2024; originally announced June 2024.

Multimode amplitude squeezing through cascaded nonlinear optical processes

Authors: Sahil Pontula, Yannick Salamin, Charles Roques-Carmes, Marin Soljacic

Abstract: Multimode squeezed light is enticing for several applications, from squeezed frequency combs for spectroscopy to signal multiplexing in optical computing. To generate squeezing in multiple frequency modes, optical parametric oscillators have been vital in realizing multimode squeezed vacuum states through second-order nonlinear processes. However, most work has focused on generating multimode sque… ▽ More Multimode squeezed light is enticing for several applications, from squeezed frequency combs for spectroscopy to signal multiplexing in optical computing. To generate squeezing in multiple frequency modes, optical parametric oscillators have been vital in realizing multimode squeezed vacuum states through second-order nonlinear processes. However, most work has focused on generating multimode squeezed vacua and squeezing in mode superpositions (supermodes). Bright squeezing in multiple discrete frequency modes, if realized, could unlock novel applications in quantum-enhanced spectroscopy and optical quantum computing. Here, we show how $Q$ factor engineering of a multimode nonlinear cavity with cascaded three wave mixing processes creates strong, spectrally tunable single mode output amplitude noise squeezing over 10 dB below the shot noise limit. In addition, we demonstrate squeezing for multiple discrete frequency modes above threshold. This bright squeezing arises from enhancement of the (noiseless) nonlinear rate relative to decay rates in the system due to the cascaded generation of photons in a single idler "bath" mode. A natural consequence of the strong nonlinear coupling in our system is the creation of an effective cavity in the synthetic frequency dimension that sustains Bloch oscillations in the modal energy distribution. Bloch mode engineering could provide an opportunity to better control nonlinear energy flow in the synthetic frequency dimension, with exciting applications in quantum random walks and topological photonics. Lastly, we show evidence of long-range correlations in amplitude noise between discrete frequency modes, pointing towards the potential of long-range entanglement in a synthetic frequency dimension. △ Less

Submitted 8 May, 2024; originally announced May 2024.

KAN: Kolmogorov-Arnold Networks

Authors: Ziming Liu, Yixuan Wang, Sachin Vaidya, Fabian Ruehle, James Halverson, Marin Soljačić, Thomas Y. Hou, Max Tegmark

Abstract: Inspired by the Kolmogorov-Arnold representation theorem, we propose Kolmogorov-Arnold Networks (KANs) as promising alternatives to Multi-Layer Perceptrons (MLPs). While MLPs have fixed activation functions on nodes ("neurons"), KANs have learnable activation functions on edges ("weights"). KANs have no linear weights at all -- every weight parameter is replaced by a univariate function parametriz… ▽ More Inspired by the Kolmogorov-Arnold representation theorem, we propose Kolmogorov-Arnold Networks (KANs) as promising alternatives to Multi-Layer Perceptrons (MLPs). While MLPs have fixed activation functions on nodes ("neurons"), KANs have learnable activation functions on edges ("weights"). KANs have no linear weights at all -- every weight parameter is replaced by a univariate function parametrized as a spline. We show that this seemingly simple change makes KANs outperform MLPs in terms of accuracy and interpretability. For accuracy, much smaller KANs can achieve comparable or better accuracy than much larger MLPs in data fitting and PDE solving. Theoretically and empirically, KANs possess faster neural scaling laws than MLPs. For interpretability, KANs can be intuitively visualized and can easily interact with human users. Through two examples in mathematics and physics, KANs are shown to be useful collaborators helping scientists (re)discover mathematical and physical laws. In summary, KANs are promising alternatives for MLPs, opening opportunities for further improving today's deep learning models which rely heavily on MLPs. △ Less

Submitted 16 June, 2024; v1 submitted 30 April, 2024; originally announced April 2024.

Comments: 48 pages, 20 figures. Codes are available at https://github.com/KindXiaoming/pykan

TENG: Time-Evolving Natural Gradient for Solving PDEs With Deep Neural Nets Toward Machine Precision

Authors: Zhuo Chen, Jacob McCarran, Esteban Vizcaino, Marin Soljačić, Di Luo

Abstract: Partial differential equations (PDEs) are instrumental for modeling dynamical systems in science and engineering. The advent of neural networks has initiated a significant shift in tackling these complexities though challenges in accuracy persist, especially for initial value problems. In this paper, we introduce the $\textit{Time-Evolving Natural Gradient (TENG)}$, generalizing time-dependent var… ▽ More Partial differential equations (PDEs) are instrumental for modeling dynamical systems in science and engineering. The advent of neural networks has initiated a significant shift in tackling these complexities though challenges in accuracy persist, especially for initial value problems. In this paper, we introduce the $\textit{Time-Evolving Natural Gradient (TENG)}$, generalizing time-dependent variational principles and optimization-based time integration, leveraging natural gradient optimization to obtain high accuracy in neural-network-based PDE solutions. Our comprehensive development includes algorithms like TENG-Euler and its high-order variants, such as TENG-Heun, tailored for enhanced precision and efficiency. TENG's effectiveness is further validated through its performance, surpassing current leading methods and achieving $\textit{machine precision}$ in step-by-step optimizations across a spectrum of PDEs, including the heat equation, Allen-Cahn equation, and Burgers' equation. △ Less

Submitted 3 June, 2024; v1 submitted 16 April, 2024; originally announced April 2024.

Report number: MIT-CTP/5706

Photonic probabilistic machine learning using quantum vacuum noise

Authors: Seou Choi, Yannick Salamin, Charles Roques-Carmes, Rumen Dangovski, Di Luo, Zhuo Chen, Michael Horodynski, Jamison Sloan, Shiekh Zia Uddin, Marin Soljacic

Abstract: Probabilistic machine learning utilizes controllable sources of randomness to encode uncertainty and enable statistical modeling. Harnessing the pure randomness of quantum vacuum noise, which stems from fluctuating electromagnetic fields, has shown promise for high speed and energy-efficient stochastic photonic elements. Nevertheless, photonic computing hardware which can control these stochastic… ▽ More Probabilistic machine learning utilizes controllable sources of randomness to encode uncertainty and enable statistical modeling. Harnessing the pure randomness of quantum vacuum noise, which stems from fluctuating electromagnetic fields, has shown promise for high speed and energy-efficient stochastic photonic elements. Nevertheless, photonic computing hardware which can control these stochastic elements to program probabilistic machine learning algorithms has been limited. Here, we implement a photonic probabilistic computer consisting of a controllable stochastic photonic element - a photonic probabilistic neuron (PPN). Our PPN is implemented in a bistable optical parametric oscillator (OPO) with vacuum-level injected bias fields. We then program a measurement-and-feedback loop for time-multiplexed PPNs with electronic processors (FPGA or GPU) to solve certain probabilistic machine learning tasks. We showcase probabilistic inference and image generation of MNIST-handwritten digits, which are representative examples of discriminative and generative models. In both implementations, quantum vacuum noise is used as a random seed to encode classification uncertainty or probabilistic generation of samples. In addition, we propose a path towards an all-optical probabilistic computing platform, with an estimated sampling rate of ~ 1 Gbps and energy consumption of ~ 5 fJ/MAC. Our work paves the way for scalable, ultrafast, and energy-efficient probabilistic machine learning hardware. △ Less

Submitted 7 March, 2024; originally announced March 2024.

Deterministic quantum state generators and stabilizers from nonlinear photonic filter cavities

Authors: Sean Chen, Nicholas Rivera, Jamison Sloan, Marin Soljacic

Abstract: Quantum states of light, particularly at optical frequencies, are considered necessary to realize a host of important quantum technologies and applications, spanning Heisenberg-limited metrology, continuous-variable quantum computing, and quantum communications. Nevertheless, a wide variety of important quantum light states are currently challenging to deterministically generate at optical frequen… ▽ More Quantum states of light, particularly at optical frequencies, are considered necessary to realize a host of important quantum technologies and applications, spanning Heisenberg-limited metrology, continuous-variable quantum computing, and quantum communications. Nevertheless, a wide variety of important quantum light states are currently challenging to deterministically generate at optical frequencies. In part, this is due to a relatively small number of schemes that prepare target quantum states given nonlinear interactions. Here, we present an especially simple concept for deterministically generating and stabilizing important quantum states of light, using only simple third-order optical nonlinearities and engineered dissipation. We show how by considering either a nonlinear cavity with frequency-dependent outcoupling, or a chain of nonlinear waveguides, one can "filter" out all but a periodic ladder of photon number components of a density matrix. As examples of this phenomenon, we show cavities which can stabilize squeezed states, as well as produce "photon-number-comb" states. Moreover, in these types of filter cavities, Glauber coherent states will deterministically evolve into Schrodinger cat states of a desired order. We discuss potential realizations in quantum nonlinear optics. More broadly, we expect that combining the techniques introduced here with additional "phase-sensitive" nonlinearities (such as second-order nonlinearity) should enable passive stabilization and generation of a wider variety of states than shown here. △ Less

Submitted 12 December, 2023; originally announced December 2023.

Multimodal Learning for Materials

Authors: Viggo Moro, Charlotte Loh, Rumen Dangovski, Ali Ghorashi, Andrew Ma, Zhuo Chen, Samuel Kim, Peter Y. Lu, Thomas Christensen, Marin Soljačić

Abstract: Artificial intelligence is transforming computational materials science, improving the prediction of material properties, and accelerating the discovery of novel materials. Recently, publicly available material data repositories have grown rapidly. This growth encompasses not only more materials, but also a greater variety and quantity of their associated properties. Existing machine learning effo… ▽ More Artificial intelligence is transforming computational materials science, improving the prediction of material properties, and accelerating the discovery of novel materials. Recently, publicly available material data repositories have grown rapidly. This growth encompasses not only more materials, but also a greater variety and quantity of their associated properties. Existing machine learning efforts in materials science focus primarily on single-modality tasks, i.e., relationships between materials and a single physical property, thus not taking advantage of the rich and multimodal set of material properties. Here, we introduce Multimodal Learning for Materials (MultiMat), which enables self-supervised multi-modality training of foundation models for materials. We demonstrate our framework's potential using data from the Materials Project database on multiple axes: (i) MultiMat achieves state-of-the-art performance for challenging material property prediction tasks; (ii) MultiMat enables novel and accurate material discovery via latent space similarity, enabling screening for stable materials with desired properties; and (iii) MultiMat encodes interpretable emergent features that may provide novel scientific insights. △ Less

Submitted 12 April, 2024; v1 submitted 30 November, 2023; originally announced December 2023.

Comments: 11 pages, 4 figures

Quantum sensitivity analysis: a general framework for controlling quantum fluctuations

Authors: Shiekh Zia Uddin, Nicholas Rivera, Devin Seyler, Yannick Salamin, Jamison Sloan, Charles Roques-Carmes, Shutao Xu, Michelle Sander, Marin Soljacic

Abstract: Nonlinear systems are important in many areas of modern science and technology. For example, nonlinearity plays an essential role in generating quantum mechanical states of both light and matter. As a result, there has been great interest in understanding the fundamental quantum nature of a variety of nonlinear effects. At the same time, there is currently a large gap between the classical and qua… ▽ More Nonlinear systems are important in many areas of modern science and technology. For example, nonlinearity plays an essential role in generating quantum mechanical states of both light and matter. As a result, there has been great interest in understanding the fundamental quantum nature of a variety of nonlinear effects. At the same time, there is currently a large gap between the classical and quantum understanding of nonlinear systems, with the classical understanding being far more developed. To close this gap, we introduce a general new theory which allows us to predict quantum effects in any nonlinear system purely in terms of its classical description. We demonstrate the predictions of our theory in experiments probing quantum fluctuations of intense femtosecond pulses propagating in an optical fiber undergoing soliton-fission supercontinuum generation, a process where broadband radiation is produced by a narrow-band input. Famously, this process is known to be highly noise-sensitive, leading to noisy outputs even from inputs with only quantum fluctuations. In contrast, our experiments uncovered a variety of previously hidden low-noise and noise-robust states arising from quantum correlations and entanglement, in agreement with the predictions of our theory. We also show how the theory points to new design concepts for controlling quantum noise in optics and beyond. We expect that our results will provide a template for discovering quantum effects in a wide variety of complex nonlinear systems. △ Less

Submitted 21 March, 2024; v1 submitted 9 November, 2023; originally announced November 2023.

Weyl points on non-orientable manifolds

Authors: André Grossi Fonseca, Sachin Vaidya, Thomas Christensen, Mikael C. Rechtsman, Taylor L. Hughes, Marin Soljačić

Abstract: Weyl fermions are hypothetical chiral particles that can also manifest as excitations near three-dimensional band crossing points in lattice systems. These quasiparticles are subject to the Nielsen-Ninomiya "no-go" theorem when placed on a lattice, requiring the total chirality across the Brillouin zone to vanish. This constraint results from the topology of the (orientable) manifold on which they… ▽ More Weyl fermions are hypothetical chiral particles that can also manifest as excitations near three-dimensional band crossing points in lattice systems. These quasiparticles are subject to the Nielsen-Ninomiya "no-go" theorem when placed on a lattice, requiring the total chirality across the Brillouin zone to vanish. This constraint results from the topology of the (orientable) manifold on which they exist. Here, we ask to what extent the concepts of topology and chirality of Weyl points remain well-defined when the underlying manifold is non-orientable. We show that the usual notion of chirality becomes ambiguous in this setting, allowing for systems with a non-zero total chirality. This circumvention of the Nielsen-Ninomiya theorem stems from a generic discontinuity of the vector field whose zeros are Weyl points. Furthermore, we discover that Weyl points on non-orientable manifolds carry an additional $\mathbb{Z}_2$ topological invariant which satisfies a different no-go theorem. We implement such Weyl points by imposing a non-symmorphic symmetry in the momentum space of lattice models. Finally, we experimentally realize all aspects of their phenomenology in a photonic platform with synthetic momenta. Our work highlights the subtle but crucial interplay between the topology of quasiparticles and of their underlying manifold. △ Less

Submitted 6 August, 2024; v1 submitted 27 October, 2023; originally announced October 2023.

Entangling extreme ultraviolet photons through strong field pair generation

Authors: Jamison Sloan, Alexey Gorlach, Matan Even Tzur, Nicholas Rivera, Oren Cohen, Ido Kaminer, Marin Soljačić

Abstract: Entangled photon pairs are a vital resource for quantum information, computation, and metrology. Although these states are routinely generated at optical frequencies, sources of quantum of light are notably lacking at extreme ultraviolet (XUV) and soft X-ray frequencies. Here, we show that strongly driven systems used for high harmonic generation (HHG) can become versatile sources of entangled pho… ▽ More Entangled photon pairs are a vital resource for quantum information, computation, and metrology. Although these states are routinely generated at optical frequencies, sources of quantum of light are notably lacking at extreme ultraviolet (XUV) and soft X-ray frequencies. Here, we show that strongly driven systems used for high harmonic generation (HHG) can become versatile sources of entangled photon pairs at these high frequencies. We present a general theory of photon pair emission from non-perturbatively driven systems, which we refer to as "strong field pair generation" (SFPG). We show that strongly driven noble gases can generate thousands of entangled pairs per shot over a large XUV bandwidth. The emitted pairs have distinctive properties in angle and frequency, which can be exploited to discriminate them from the background HHG signal. We connect SFPG theory to the three-step-model of HHG, showing that this pair emission originates from the impact of high frequency vacuum fluctuations on electron recombination. The light produced by SFPG exhibits attosecond Hong-Ou-Mandel correlations, and can be leveraged as a source of heralded single photon attosecond pulses. Our findings aid ongoing efforts to propel quantum optics into the XUV and beyond. △ Less

Submitted 28 September, 2023; originally announced September 2023.

Driven-dissipative phases and dynamics in non-Markovian nonlinear photonics

Authors: Jamison Sloan, Nicholas Rivera, Marin Soljačić

Abstract: Interactions between photons (nonlinearities) enable a powerful form of control over the state of light. This control has enabled technologies such as light sources at new wavelengths, ultra-short optical pulses, frequency-comb metrology systems, even quantum light sources. Common to a wide variety of nonlinear optical technologies is an equilibrium between an energy source, such as an external la… ▽ More Interactions between photons (nonlinearities) enable a powerful form of control over the state of light. This control has enabled technologies such as light sources at new wavelengths, ultra-short optical pulses, frequency-comb metrology systems, even quantum light sources. Common to a wide variety of nonlinear optical technologies is an equilibrium between an energy source, such as an external laser, and dissipation, such as radiation loss or absorption. In the vast majority of these systems, the coupling between the system and the outside world (which leads to loss) is well-described as ``Markovian,'' meaning that the outside world has no memory of its past state. In this work, we introduce a class of driven-dissipative systems in which a nonlinear cavity experiences non-Markovian coupling to the outside world. In the classical regime, we show that these non-Markovian cavities can have extremely low thresholds for nonlinear effects, as well as self-pulsing instabilities at THz rates, and rich phase diagrams with alternating regions of stability and instability. In the quantum regime, we show how these system, when implemented on state-of-the-art platforms, can enable generation of strongly squeezed cavity states with intensity fluctuations that can be more than 15 dB below the classical limit, in contrast to the Markovian driven-dissipative cavity, in which the limit is 3 dB. In the regime of few-photon nonlinearity, such non-Markovian cavities can enable a deterministic protocol to generate Fock states of high order, which are long-desired, but still elusive at optical frequencies. We expect that exploiting non-Markovian couplings in nonlinear optics should in the future lead to even richer possibilities than those discussed here for both classical and quantum light manipulations. △ Less

Submitted 18 September, 2023; originally announced September 2023.

Prevalence of two-dimensional photonic topology

Authors: Ali Ghorashi, Sachin Vaidya, Mikael Rechtsman, Wladimir Benalcazar, Marin Soljačić, Thomas Christensen

Abstract: The topological characteristics of photonic crystals have been the subject of intense research in recent years. Despite this, the basic question of whether photonic band topology is rare or abundant -- i.e., its relative prevalence -- remains unaddressed. Here, we determine the prevalence of stable, fragile, and higher-order photonic topology in the 11 two-dimensional crystallographic symmetry set… ▽ More The topological characteristics of photonic crystals have been the subject of intense research in recent years. Despite this, the basic question of whether photonic band topology is rare or abundant -- i.e., its relative prevalence -- remains unaddressed. Here, we determine the prevalence of stable, fragile, and higher-order photonic topology in the 11 two-dimensional crystallographic symmetry settings that admit diagnosis of one or more of these phenomena by symmetry analysis. Our determination is performed on the basis of a data set of 550000 randomly sampled, two-tone photonic crystals, spanning 11 symmetry settings and 5 dielectric contrasts, and examined in both transverse electric (TE) and magnetic (TM) polarizations. We report the abundance of nontrivial photonic topology in the presence of time-reversal symmetry and find that stable, fragile, and higher-order topology are generally abundant. Below the first band gap, which is of primary experimental interest, we find that stable topology is more prevalent in the TE polarization than the TM; is only weakly, but monotonically, dependent on dielectric contrast; and that fragile topology is near-absent. In the absence of time-reversal symmetry, nontrivial Chern phases are also abundant in photonic crystals with 2-, 4-, and 6-fold rotational symmetries but comparatively rare in settings with only 3-fold symmetry. Our results elucidate the interplay of symmetry, dielectric contrast, electromagnetic polarization, and time-reversal breaking in engendering topological photonic phases and may inform general design principles for their experimental realization. △ Less

Submitted 28 July, 2023; originally announced July 2023.

Intense squeezed light from lasers with sharply nonlinear gain at optical frequencies

Authors: Linh Nguyen, Jamison Sloan, Nicholas Rivera, Marin Soljacic

Abstract: Non-classical states of light, such as number-squeezed light, with fluctuations below the classical shot noise level, have important uses in metrology, communication, quantum information processing, and quantum simulation. However, generating these non-classical states of light, especially with high intensity and high degree of squeezing, is challenging. To address this problem, we introduce a new… ▽ More Non-classical states of light, such as number-squeezed light, with fluctuations below the classical shot noise level, have important uses in metrology, communication, quantum information processing, and quantum simulation. However, generating these non-classical states of light, especially with high intensity and high degree of squeezing, is challenging. To address this problem, we introduce a new concept which uses gain to generate intense sub-Poissonian light at optical frequencies. It exploits a strongly nonlinear gain for photons which arises from a combination of frequency-dependent gain and Kerr nonlinearity. In this laser architecture, the interaction between the gain medium and Kerr nonlinearity suppresses the spontaneous emission at high photon number states, leading to a strong "negative feedback" that suppresses photon-number fluctuations. We discuss realistic implementations of this concept based on the use of solid-state gain media in laser cavities with Kerr nonlinear materials, showing how 90% squeezing of photon number fluctuations below the shot noise level can be realized. △ Less

Submitted 2 June, 2023; originally announced June 2023.

Highly-confined and tunable plasmonics based on two-dimensional solid-state defect lattices

Authors: Ali Ghorashi, Nicholas Rivera, Bowen Shi, Ravishankar Sundararaman, Efthimios Kaxiras, John Joannopoulos, Marin Soljacic

Abstract: Plasmons, collective excitations of electrons in solids, are associated with strongly confined electromagnetic fields, with wavelengths far below the wavelength of photons in free space. This strong confinement promises the realization of optoelectronic devices that could bridge the size difference between photonic and electronic devices. However, despite decades of research in plasmonics, many ap… ▽ More Plasmons, collective excitations of electrons in solids, are associated with strongly confined electromagnetic fields, with wavelengths far below the wavelength of photons in free space. This strong confinement promises the realization of optoelectronic devices that could bridge the size difference between photonic and electronic devices. However, despite decades of research in plasmonics, many applications remain limited by plasmonic losses, thus motivating a search for new engineered plasmonic materials with lower losses. A promising pathway for low-loss plasmonic materials is the engineering of materials with flat and energetically isolated metallic bands, which can strongly limit phonon-assisted optical losses, a major contributor to short plasmonic lifetimes. Such electronic band structures may be created by judiciously introducing an ordered lattice of defects in an insulating host material. Here, we explore this approach, presenting several low-loss, highly-confined, and tunable plasmonic materials based on arrays of carbon substitutions in hexagonal boron nitride (hBN) monolayers. From our first-principles calculations based on density functional theory (DFT), we find plasmonic structures with mid-infrared plasmons featuring very high confinements ($λ_{\text{vacuum}}/λ_{\text{plasmon}}$ exceeding 2000) and quality factors in excess of 1000. We provide a systematic explanation of how crystal structure, electronic bandwidth, and many-body effects affect the plasmonic dispersions and losses of these materials. The results are thus of relevance to low-loss plasmon engineering in other flat band systems. △ Less

Submitted 1 May, 2023; originally announced May 2023.

ANTN: Bridging Autoregressive Neural Networks and Tensor Networks for Quantum Many-Body Simulation

Authors: Zhuo Chen, Laker Newhouse, Eddie Chen, Di Luo, Marin Soljačić

Abstract: Quantum many-body physics simulation has important impacts on understanding fundamental science and has applications to quantum materials design and quantum technology. However, due to the exponentially growing size of the Hilbert space with respect to the particle number, a direct simulation is intractable. While representing quantum states with tensor networks and neural networks are the two sta… ▽ More Quantum many-body physics simulation has important impacts on understanding fundamental science and has applications to quantum materials design and quantum technology. However, due to the exponentially growing size of the Hilbert space with respect to the particle number, a direct simulation is intractable. While representing quantum states with tensor networks and neural networks are the two state-of-the-art methods for approximate simulations, each has its own limitations in terms of expressivity and inductive bias. To address these challenges, we develop a novel architecture, Autoregressive Neural TensorNet (ANTN), which bridges tensor networks and autoregressive neural networks. We show that Autoregressive Neural TensorNet parameterizes normalized wavefunctions, allows for exact sampling, generalizes the expressivity of tensor networks and autoregressive neural networks, and inherits a variety of symmetries from autoregressive neural networks. We demonstrate our approach on quantum state learning as well as finding the ground state of the challenging 2D $J_1$-$J_2$ Heisenberg model with different systems sizes and coupling parameters, outperforming both tensor networks and autoregressive neural networks. Our work opens up new opportunities for quantum many-body physics simulation, quantum technology design, and generative modeling in artificial intelligence. △ Less

Submitted 16 April, 2024; v1 submitted 4 April, 2023; originally announced April 2023.

Report number: MIT-CTP/5549

Model Stitching: Looking For Functional Similarity Between Representations

Authors: Adriano Hernandez, Rumen Dangovski, Peter Y. Lu, Marin Soljacic

Abstract: Model stitching (Lenc & Vedaldi 2015) is a compelling methodology to compare different neural network representations, because it allows us to measure to what degree they may be interchanged. We expand on a previous work from Bansal, Nakkiran & Barak which used model stitching to compare representations of the same shapes learned by differently seeded and/or trained neural networks of the same arc… ▽ More Model stitching (Lenc & Vedaldi 2015) is a compelling methodology to compare different neural network representations, because it allows us to measure to what degree they may be interchanged. We expand on a previous work from Bansal, Nakkiran & Barak which used model stitching to compare representations of the same shapes learned by differently seeded and/or trained neural networks of the same architecture. Our contribution enables us to compare the representations learned by layers with different shapes from neural networks with different architectures. We subsequently reveal unexpected behavior of model stitching. Namely, we find that stitching, based on convolutions, for small ResNets, can reach high accuracy if those layers come later in the first (sender) network than in the second (receiver), even if those layers are far apart. △ Less

Submitted 31 August, 2023; v1 submitted 20 March, 2023; originally announced March 2023.

Comments: 5 pages, 2 figures

Topological electromagnetic waves in dispersive and lossy plasma crystals

Authors: Chen Qian, Yue Jiang, Jicheng Jin, Thomas Christensen, Marin Soljačić, Alexander V. Kildishev, Bo Zhen

Abstract: Topological photonic crystals, which offer topologically protected and back-scattering-immune transport channels, have recently gained significant attention for both scientific and practical reasons. Although most current studies focus on dielectric materials with weak dispersions, this study focuses on topological phases in dispersive materials and presents a numerical study of Chern insulators i… ▽ More Topological photonic crystals, which offer topologically protected and back-scattering-immune transport channels, have recently gained significant attention for both scientific and practical reasons. Although most current studies focus on dielectric materials with weak dispersions, this study focuses on topological phases in dispersive materials and presents a numerical study of Chern insulators in gaseous-phase plasma cylinder cells. We develop a numerical framework to address the complex material dispersion arising from the plasma medium and external magnetic fields and identify Chern insulator phases that are experimentally achievable. Using this numerical tool, we also explain the flat bands commonly observed in periodic plasmonic structures, via local resonances, and how edge states change as the edge termination is periodically modified. This work opens up opportunities for exploring band topology in new materials with non-trivial dispersions and has potential RF applications, ranging from plasma-based lighting to plasma propulsion engines. △ Less

Submitted 11 March, 2023; v1 submitted 8 March, 2023; originally announced March 2023.

Comments: 10 pages, 4 figures

Biasing the quantum vacuum to control macroscopic probability distributions

Authors: Charles Roques-Carmes, Yannick Salamin, Jamison Sloan, Seou Choi, Gustavo Velez, Ethan Koskas, Nicholas Rivera, Steven E. Kooi, John D. Joannopoulos, Marin Soljacic

Abstract: One of the most important insights of quantum field theory is that electromagnetic fields must fluctuate. Even in the vacuum state, the electric and magnetic fields have a nonzero variance, leading to ubiquitous effects such as spontaneous emission, the Lamb shift, the Casimir effect, and more. These "vacuum fluctuations" have also been harnessed as a source of perfect randomness, for example to g… ▽ More One of the most important insights of quantum field theory is that electromagnetic fields must fluctuate. Even in the vacuum state, the electric and magnetic fields have a nonzero variance, leading to ubiquitous effects such as spontaneous emission, the Lamb shift, the Casimir effect, and more. These "vacuum fluctuations" have also been harnessed as a source of perfect randomness, for example to generate perfectly random photonic bits. Despite these achievements, many potential applications of quantum randomness in fields such as probabilistic computing rely on controllable probability distributions, which have not yet been realized on photonic platforms. In this work, we show that the injection of vacuum-level "bias" fields into a multi-stable optical system enables a controllable source of "biased" quantum randomness. We demonstrate this concept in an optical parametric oscillator (OPO). Ordinarily, an OPO initiated from the ground state develops a signal field in one of two degenerate phase states (0 and $π$) with equal probability. By injecting bias pulses which contain less than one photon on average, we control the probabilities associated with the two output states, leading to the first controllable photonic probabilistic bit (p-bit). We shed light on the physics behind this process, showing quantitative agreement between theory and experiment. Finally, we demonstrate the potential of our approach for sensing sub-photon level fields by showing that our system is sensitive to the temporal shape of bias field pulses far below the single photon level. Our results suggest a new platform for the study of stochastic quantum dynamics in nonlinear driven-dissipative systems, and point toward possible applications in ultrafast photonic probabilistic computing, as well as the sensing of extremely weak fields. △ Less

Submitted 24 April, 2023; v1 submitted 6 March, 2023; originally announced March 2023.

Multi-Symmetry Ensembles: Improving Diversity and Generalization via Opposing Symmetries

Authors: Charlotte Loh, Seungwook Han, Shivchander Sudalairaj, Rumen Dangovski, Kai Xu, Florian Wenzel, Marin Soljacic, Akash Srivastava

Abstract: Deep ensembles (DE) have been successful in improving model performance by learning diverse members via the stochasticity of random initialization. While recent works have attempted to promote further diversity in DE via hyperparameters or regularizing loss functions, these methods primarily still rely on a stochastic approach to explore the hypothesis space. In this work, we present Multi-Symmetr… ▽ More Deep ensembles (DE) have been successful in improving model performance by learning diverse members via the stochasticity of random initialization. While recent works have attempted to promote further diversity in DE via hyperparameters or regularizing loss functions, these methods primarily still rely on a stochastic approach to explore the hypothesis space. In this work, we present Multi-Symmetry Ensembles (MSE), a framework for constructing diverse ensembles by capturing the multiplicity of hypotheses along symmetry axes, which explore the hypothesis space beyond stochastic perturbations of model weights and hyperparameters. We leverage recent advances in contrastive representation learning to create models that separately capture opposing hypotheses of invariant and equivariant functional classes and present a simple ensembling approach to efficiently combine appropriate hypotheses for a given task. We show that MSE effectively captures the multiplicity of conflicting hypotheses that is often required in large, diverse datasets like ImageNet. As a result of their inherent diversity, MSE improves classification performance, uncertainty quantification, and generalization across a series of transfer tasks. △ Less

Submitted 19 June, 2023; v1 submitted 4 March, 2023; originally announced March 2023.

Comments: Camera Ready Revision. ICML 2023

Q-Flow: Generative Modeling for Differential Equations of Open Quantum Dynamics with Normalizing Flows

Authors: Owen Dugan, Peter Y. Lu, Rumen Dangovski, Di Luo, Marin Soljačić

Abstract: Studying the dynamics of open quantum systems can enable breakthroughs both in fundamental physics and applications to quantum engineering and quantum computation. Since the density matrix $ρ$, which is the fundamental description for the dynamics of such systems, is high-dimensional, customized deep generative neural networks have been instrumental in modeling $ρ$. However, the complex-valued nat… ▽ More Studying the dynamics of open quantum systems can enable breakthroughs both in fundamental physics and applications to quantum engineering and quantum computation. Since the density matrix $ρ$, which is the fundamental description for the dynamics of such systems, is high-dimensional, customized deep generative neural networks have been instrumental in modeling $ρ$. However, the complex-valued nature and normalization constraints of $ρ$, as well as its complicated dynamics, prohibit a seamless connection between open quantum systems and the recent advances in deep generative modeling. Here we lift that limitation by utilizing a reformulation of open quantum system dynamics to a partial differential equation (PDE) for a corresponding probability distribution $Q$, the Husimi Q function. Thus, we model the Q function seamlessly with off-the-shelf deep generative models such as normalizing flows. Additionally, we develop novel methods for learning normalizing flow evolution governed by high-dimensional PDEs based on the Euler method and the application of the time-dependent variational principle. We name the resulting approach $Q$-$Flow$ and demonstrate the scalability and efficiency of Q-Flow on open quantum system simulations, including the dissipative harmonic oscillator and the dissipative bosonic model. Q-Flow is superior to conventional PDE solvers and state-of-the-art physics-informed neural network solvers, especially in high-dimensional systems. △ Less

Submitted 6 June, 2023; v1 submitted 23 February, 2023; originally announced February 2023.

Report number: MIT-CTP/5533

Geometry of contact: contact planning for multi-legged robots via spin models duality

Authors: Baxi Chong, Di Luo, Tianyu Wang, Gabriel Margolis, Juntao He, Pulkit Agrawal, Marin Soljačić, Daniel I. Goldman

Abstract: Contact planning is crucial in locomoting systems.Specifically, appropriate contact planning can enable versatile behaviors (e.g., sidewinding in limbless locomotors) and facilitate speed-dependent gait transitions (e.g., walk-trot-gallop in quadrupedal locomotors). The challenges of contact planning include determining not only the sequence by which contact is made and broken between the locomoto… ▽ More Contact planning is crucial in locomoting systems.Specifically, appropriate contact planning can enable versatile behaviors (e.g., sidewinding in limbless locomotors) and facilitate speed-dependent gait transitions (e.g., walk-trot-gallop in quadrupedal locomotors). The challenges of contact planning include determining not only the sequence by which contact is made and broken between the locomotor and the environments, but also the sequence of internal shape changes (e.g., body bending and limb shoulder joint oscillation). Most state-of-art contact planning algorithms focused on conventional robots (e.g.biped and quadruped) and conventional tasks (e.g. forward locomotion), and there is a lack of study on general contact planning in multi-legged robots. In this paper, we show that using geometric mechanics framework, we can obtain the global optimal contact sequence given the internal shape changes sequence. Therefore, we simplify the contact planning problem to a graph optimization problem to identify the internal shape changes. Taking advantages of the spatio-temporal symmetry in locomotion, we map the graph optimization problem to special cases of spin models, which allows us to obtain the global optima in polynomial time. We apply our approach to develop new forward and sidewinding behaviors in a hexapod and a 12-legged centipede. We verify our predictions using numerical and robophysical models, and obtain novel and effective locomotion behaviors. △ Less

Submitted 7 February, 2023; v1 submitted 6 February, 2023; originally announced February 2023.

Comments: SI video: https://doi.org/10.5281/zenodo.7608693

Report number: MIT-CTP/5526

Transcending shift-invariance in the paraxial regime via end-to-end inverse design of freeform nanophotonics

Authors: William F. Li, Gaurav Arya, Charles Roques-Carmes, Zin Lin, Steven G. Johnson, Marin Soljačić

Abstract: Traditional optical elements and conventional metasurfaces obey shift-invariance in the paraxial regime. For imaging systems obeying paraxial shift-invariance, a small shift in input angle causes a corresponding shift in the sensor image. Shift-invariance has deep implications for the design and functionality of optical devices, such as the necessity of free space between components (as in compoun… ▽ More Traditional optical elements and conventional metasurfaces obey shift-invariance in the paraxial regime. For imaging systems obeying paraxial shift-invariance, a small shift in input angle causes a corresponding shift in the sensor image. Shift-invariance has deep implications for the design and functionality of optical devices, such as the necessity of free space between components (as in compound objectives made of several curved surfaces). We present a method for nanophotonic inverse design of compact imaging systems whose resolution is not constrained by paraxial shift-invariance. Our method is end-to-end, in that it integrates density-based full-Maxwell topology optimization with a fully iterative elastic-net reconstruction algorithm. By the design of nanophotonic structures that scatter light in a non-shift-invariant manner, our optimized nanophotonic imaging system overcomes the limitations of paraxial shift-invariance, achieving accurate, noise-robust image reconstruction beyond shift-invariant resolution. △ Less

Submitted 3 February, 2023; originally announced February 2023.

Journal ref: Optics Express 31 (2023) 24260-24272

Strong broadband intensity noise squeezing from near-infrared to terahertz frequencies in semiconductor lasers with nonlinear dissipation

Authors: Sahil Pontula, Jamison Sloan, Nicholas Rivera, Marin Soljacic

Abstract: The generation and application of squeezed light have long been central goals of quantum optics, enabling sensing below the standard quantum limit, optical quantum computing platforms, and more. Intensity noise squeezing of bright (coherent) states, in contrast to squeezed vacuum, is relatively underdeveloped. Bright squeezing has been generated directly through nonlinear optical processes or ``qu… ▽ More The generation and application of squeezed light have long been central goals of quantum optics, enabling sensing below the standard quantum limit, optical quantum computing platforms, and more. Intensity noise squeezing of bright (coherent) states, in contrast to squeezed vacuum, is relatively underdeveloped. Bright squeezing has been generated directly through nonlinear optical processes or ``quietly pumped'' semiconductor lasers. However, these methods suffer from weak squeezing limits, narrow operating wavelength ranges, and have not been explored at large bandwidths. Here, we show how semiconductor lasers with sharp intensity-dependent dissipation can support highly broadband intensity noise squeezing from infrared (IR) to terahertz (THz) wavelengths, the latter of which has remained unexplored in quantum noise studies. Our protocol realizes strongly ($>10$ dB) intensity noise-squeezed intracavity quantum states, which could create a new regime for cavity quantum electrodynamics experiments, as well as strong output squeezing surpassing gigahertz bandwidths. Furthermore, we show how the same systems also create self-pulsing and bistable mean field behavior, enabling control of light in both the temporal and noise domains. The existence of these multiple functionalities in both the classical and quantum mechanical domains in a single semiconductor laser platform, from IR to THz wavelengths, could enable advances in on-chip quantum optical communication, computing, and sensing across the electromagnetic spectrum. △ Less

Submitted 7 May, 2024; v1 submitted 14 December, 2022; originally announced December 2022.

Automated Discovery and Optimization of 3D Topological Photonic Crystals

Authors: Samuel Kim, Thomas Christensen, Steven G. Johnson, Marin Soljačić

Abstract: Topological photonic crystals have received considerable attention for their ability to manipulate and guide light in unique ways. They are typically designed by hand based on careful analysis of their bands and mode profiles, but recent theoretical advances have revealed new and powerful insights into the connection between band symmetry, connectivity, and topology. Here we propose a combined glo… ▽ More Topological photonic crystals have received considerable attention for their ability to manipulate and guide light in unique ways. They are typically designed by hand based on careful analysis of their bands and mode profiles, but recent theoretical advances have revealed new and powerful insights into the connection between band symmetry, connectivity, and topology. Here we propose a combined global and local optimization framework that integrates a flexible symmetry-constrained level-set parameterization with standard gradient-free optimization algorithms to optimize topological photonic crystals, a problem setting where the objective function may be highly non-convex and non-continuous. Our framework can be applied to any symmetry-identifiable band topology, and we demonstrate its applicability to several prominent kinds of three-dimensional band topology, namely $Γ$-enforced nodal lines, Weyl points, and Chern insulators. Requiring no prior examples of topological photonic crystals or prior knowledge on the connection between structure and band topology, our approach indicates a path towards the automated discovery of novel topological photonic crystal designs. △ Less

Submitted 29 November, 2022; originally announced November 2022.

Comments: 14 pages, 7 figures

Optical properties of dispersive time-dependent materials

Authors: Jamison Sloan, Nicholas Rivera, John D Joannopoulos, Marin Soljačić

Abstract: Time-varying optical materials have attracted recent interest for their potential to enable frequency conversion, nonreciprocal physics, photonic time-crystals, and more. However, the description of time-varying materials has been primarily limited to regimes where material resonances (i.e., dispersion) can be neglected. In this work, we describe how the optics of these dispersive time-varying mat… ▽ More Time-varying optical materials have attracted recent interest for their potential to enable frequency conversion, nonreciprocal physics, photonic time-crystals, and more. However, the description of time-varying materials has been primarily limited to regimes where material resonances (i.e., dispersion) can be neglected. In this work, we describe how the optics of these dispersive time-varying materials emerges from microscopic quantum mechanical models of time-driven systems. Our results are based on a framework for describing the optics of dispersive time-varying materials through quantum mechanical linear response theory. Importantly, we clarify how response functions for time-varying materials are connected to energy transfer. We provide three examples of our framework applied to systems which can be used to model a wide variety of experiments: few level models that can describe atoms, spins, or superconducting qubits, oscillator models which can describe the strong response of polar insulators, and strongly driven atom models which can describe the highly nonperturbative optical response of materials undergoing high harmonic generation. We anticipate that our results will be broadly applicable to electromagnetic phenomena in strongly time-varying systems. △ Less

Submitted 29 November, 2022; originally announced November 2022.

Quasicrystalline Weyl points and dense Fermi-Bragg arcs

Authors: André Grossi e Fonseca, Thomas Christensen, John D. Joannopoulos, Marin Soljačić

Abstract: We introduce a general mechanism for obtaining Weyl points in a stack of 2D quasicrystals, which can be extended to any stack of aperiodic layers. We do so by driving a topological phase transition with the vertical crystal-momentum as the tuning parameter, which leads to gap closures at the critical points sourcing Berry curvature. To illustrate, we use a simple 3D generalization of the Qi-Wu-Zha… ▽ More We introduce a general mechanism for obtaining Weyl points in a stack of 2D quasicrystals, which can be extended to any stack of aperiodic layers. We do so by driving a topological phase transition with the vertical crystal-momentum as the tuning parameter, which leads to gap closures at the critical points sourcing Berry curvature. To illustrate, we use a simple 3D generalization of the Qi-Wu-Zhang model defined on a Penrose quasicrystal. The presence of Weyl points is established via the local Chern marker, projected band structure and density of states. Interestingly, we uncover an analogue of Fermi arcs in the quasicrystalline setting, which we deem Fermi-Bragg arcs, densely distributed lines connecting the band degeneracies and indexed by the Bragg peaks. Signatures of such surface states in quantum oscillations and the prospect of a fully quasicrystalline Weyl system are also discussed. The flexibility of our proposal brings new opportunities for realizing other gapless topological phases in aperiodic systems, paving the way for a significantly expanded role for topological band theory. △ Less

Submitted 16 August, 2023; v1 submitted 25 November, 2022; originally announced November 2022.

Comments: 6 pages, 3 figures

Journal ref: Phys. Rev. B 108, L121109 (2023)

Creating large Fock states and massively squeezed states in optics using systems with nonlinear bound states in the continuum

Authors: Nicholas Rivera, Jamison Sloan, Yannick Salamin, John D. Joannopoulos, Marin Soljacic

Abstract: The quantization of the electromagnetic field leads directly to the existence of quantum mechanical states, called Fock states, with an exact integer number of photons. Despite these fundamental states being long-understood, and despite their many potential applications, generating them is largely an open problem. For example, at optical frequencies, it is challenging to deterministically generate… ▽ More The quantization of the electromagnetic field leads directly to the existence of quantum mechanical states, called Fock states, with an exact integer number of photons. Despite these fundamental states being long-understood, and despite their many potential applications, generating them is largely an open problem. For example, at optical frequencies, it is challenging to deterministically generate Fock states of order two and beyond. Here, we predict the existence of a new effect in nonlinear optics, which enables the deterministic generation of large Fock states at arbitrary frequencies. The effect, which we call an n-photon bound state in the continuum, is one in which a photonic resonance (such as a cavity mode) becomes lossless when a precise number of photons n is inside the resonance. Based on analytical theory and numerical simulations, we show that these bound states enable a remarkable phenomenon in which a coherent state of light, when injected into a system supporting this bound state, can spontaneously evolve into a Fock state of a controllable photon number. This effect is also directly applicable for creating (highly) squeezed states of light, whose photon number fluctuations are (far) below the value expected from classical physics (i.e., shot noise). We suggest several examples of systems to experimentally realize the effects predicted here in nonlinear nanophotonic systems, showing examples of generating both optical Fock states with large n (n > 10), as well as more macroscopic photonic states with very large squeezing, with over 90% less noise (10 dB) than the classical value associated with shot noise. △ Less

Submitted 10 November, 2022; v1 submitted 2 November, 2022; originally announced November 2022.

QuACK: Accelerating Gradient-Based Quantum Optimization with Koopman Operator Learning

Authors: Di Luo, Jiayu Shen, Rumen Dangovski, Marin Soljačić

Abstract: Quantum optimization, a key application of quantum computing, has traditionally been stymied by the linearly increasing complexity of gradient calculations with an increasing number of parameters. This work bridges the gap between Koopman operator theory, which has found utility in applications because it allows for a linear representation of nonlinear dynamical systems, and natural gradient metho… ▽ More Quantum optimization, a key application of quantum computing, has traditionally been stymied by the linearly increasing complexity of gradient calculations with an increasing number of parameters. This work bridges the gap between Koopman operator theory, which has found utility in applications because it allows for a linear representation of nonlinear dynamical systems, and natural gradient methods in quantum optimization, leading to a significant acceleration of gradient-based quantum optimization. We present Quantum-circuit Alternating Controlled Koopman learning (QuACK), a novel framework that leverages an alternating algorithm for efficient prediction of gradient dynamics on quantum computers. We demonstrate QuACK's remarkable ability to accelerate gradient-based optimization across a range of applications in quantum optimization and machine learning. In fact, our empirical studies, spanning quantum chemistry, quantum condensed matter, quantum machine learning, and noisy environments, have shown accelerations of more than 200x speedup in the overparameterized regime, 10x speedup in the smooth regime, and 3x speedup in the non-smooth regime. With QuACK, we offer a robust advancement that harnesses the advantage of gradient-based quantum optimization for practical benefits. △ Less

Submitted 4 May, 2024; v1 submitted 2 November, 2022; originally announced November 2022.

Comments: Advances in Neural Information Processing Systems 36 (NeurIPS 2023) spotlight

Report number: MIT-CTP/5488

Learning to Optimize Quasi-Newton Methods

Authors: Isaac Liao, Rumen R. Dangovski, Jakob N. Foerster, Marin Soljačić

Abstract: Fast gradient-based optimization algorithms have become increasingly essential for the computationally efficient training of machine learning models. One technique is to multiply the gradient by a preconditioner matrix to produce a step, but it is unclear what the best preconditioner matrix is. This paper introduces a novel machine learning optimizer called LODO, which tries to online meta-learn t… ▽ More Fast gradient-based optimization algorithms have become increasingly essential for the computationally efficient training of machine learning models. One technique is to multiply the gradient by a preconditioner matrix to produce a step, but it is unclear what the best preconditioner matrix is. This paper introduces a novel machine learning optimizer called LODO, which tries to online meta-learn the best preconditioner during optimization. Specifically, our optimizer merges Learning to Optimize (L2O) techniques with quasi-Newton methods to learn preconditioners parameterized as neural networks; they are more flexible than preconditioners in other quasi-Newton methods. Unlike other L2O methods, LODO does not require any meta-training on a training task distribution, and instead learns to optimize on the fly while optimizing on the test task, adapting to the local characteristics of the loss landscape while traversing it. Theoretically, we show that our optimizer approximates the inverse Hessian in noisy loss landscapes and is capable of representing a wide range of inverse Hessians. We experimentally verify that our algorithm can optimize in noisy settings, and show that simpler alternatives for representing the inverse Hessians worsen performance. Lastly, we use our optimizer to train a semi-realistic deep neural network with 95k parameters at speeds comparable to those of standard neural network optimizers. △ Less

Submitted 11 September, 2023; v1 submitted 10 October, 2022; originally announced October 2022.

ACM Class: I.2.6

On the Importance of Calibration in Semi-supervised Learning

Authors: Charlotte Loh, Rumen Dangovski, Shivchander Sudalairaj, Seungwook Han, Ligong Han, Leonid Karlinsky, Marin Soljacic, Akash Srivastava

Abstract: State-of-the-art (SOTA) semi-supervised learning (SSL) methods have been highly successful in leveraging a mix of labeled and unlabeled data by combining techniques of consistency regularization and pseudo-labeling. During pseudo-labeling, the model's predictions on unlabeled data are used for training and thus, model calibration is important in mitigating confirmation bias. Yet, many SOTA methods… ▽ More State-of-the-art (SOTA) semi-supervised learning (SSL) methods have been highly successful in leveraging a mix of labeled and unlabeled data by combining techniques of consistency regularization and pseudo-labeling. During pseudo-labeling, the model's predictions on unlabeled data are used for training and thus, model calibration is important in mitigating confirmation bias. Yet, many SOTA methods are optimized for model performance, with little focus directed to improve model calibration. In this work, we empirically demonstrate that model calibration is strongly correlated with model performance and propose to improve calibration via approximate Bayesian techniques. We introduce a family of new SSL models that optimizes for calibration and demonstrate their effectiveness across standard vision benchmarks of CIFAR-10, CIFAR-100 and ImageNet, giving up to 15.9% improvement in test accuracy. Furthermore, we also demonstrate their effectiveness in additional realistic and challenging problems, such as class-imbalanced datasets and in photonics science. △ Less

Submitted 10 October, 2022; originally announced October 2022.

Comments: 24 pages

AI-Assisted Discovery of Quantitative and Formal Models in Social Science

Authors: Julia Balla, Sihao Huang, Owen Dugan, Rumen Dangovski, Marin Soljacic

Abstract: In social science, formal and quantitative models, such as ones describing economic growth and collective action, are used to formulate mechanistic explanations, provide predictions, and uncover questions about observed phenomena. Here, we demonstrate the use of a machine learning system to aid the discovery of symbolic models that capture nonlinear and dynamical relationships in social science da… ▽ More In social science, formal and quantitative models, such as ones describing economic growth and collective action, are used to formulate mechanistic explanations, provide predictions, and uncover questions about observed phenomena. Here, we demonstrate the use of a machine learning system to aid the discovery of symbolic models that capture nonlinear and dynamical relationships in social science datasets. By extending neuro-symbolic methods to find compact functions and differential equations in noisy and longitudinal data, we show that our system can be used to discover interpretable models from real-world data in economics and sociology. Augmenting existing workflows with symbolic regression can help uncover novel relationships and explore counterfactual models during the scientific process. We propose that this AI-assisted framework can bridge parametric and non-parametric models commonly employed in social science research by systematically exploring the space of nonlinear models and enabling fine-grained control over expressivity and interpretability. △ Less

Submitted 16 August, 2023; v1 submitted 2 October, 2022; originally announced October 2022.

Comments: 19 pages, 4 figures

Non-Abelian nonsymmorphic chiral symmetries

Authors: Yi Yang, Hoi Chun Po, Vincent Liu, John D. Joannopoulos, Liang Fu, Marin Soljačić

Abstract: The Hofstadter model exemplifies a large class of physical systems characterized by particles hopping on a lattice immersed in a gauge field. Recent advancements on various synthetic platforms have enabled highly-controllable simulations of such systems with tailored gauge fields featuring complex spatial textures. These synthetic gauge fields could introduce synthetic symmetries that do not appea… ▽ More The Hofstadter model exemplifies a large class of physical systems characterized by particles hopping on a lattice immersed in a gauge field. Recent advancements on various synthetic platforms have enabled highly-controllable simulations of such systems with tailored gauge fields featuring complex spatial textures. These synthetic gauge fields could introduce synthetic symmetries that do not appear in electronic materials. Here, in an SU(2) non-Abelian Hofstadter model, we theoretically show the emergence of multiple nonsymmorphic chiral symmetries, which combine an internal unitary anti-symmetry with fractional spatial translation. Depending on the values of the gauge fields, the nonsymmorphic chiral symmetries can exhibit non-Abelian algebra and protect Kramer quartet states in the bulk band structure, creating general four-fold degeneracy at all momenta. These nonsymmorphic chiral symmetries protect double Dirac semimetals at zero energy, which become gapped into quantum confined insulating phases upon introducing a boundary. Moreover, the parity of the system size can determine whether the resulting insulating phase is trivial or topological. Our work indicates a pathway for creating topology via synthetic symmetries emergent from synthetic gauge fields. △ Less

Submitted 14 September, 2022; originally announced September 2022.

Quantum Electrodynamical Metamaterials

Authors: Josephine Yu, Jamison Sloan, Nicholas Rivera, Marin Soljacic

Abstract: Recent experiments have revealed ultrastrong coupling between light and matter as a promising avenue for modifying material properties, such as electrical transport, chemical reaction rates, and even superconductivity. Here, we explore (ultra)strong coupling as a means for manipulating the optical response of metamaterials based on ensembles of constituent units individually in the ultrastrong cou… ▽ More Recent experiments have revealed ultrastrong coupling between light and matter as a promising avenue for modifying material properties, such as electrical transport, chemical reaction rates, and even superconductivity. Here, we explore (ultra)strong coupling as a means for manipulating the optical response of metamaterials based on ensembles of constituent units individually in the ultrastrong coupling regime. We develop a framework based on linear response for quantum electrodynamical systems to study how light-matter coupling affects the optical response. We begin by applying this framework to find the optical response of a two-level emitter coupled to a single cavity mode, which could be seen as a "meta-atom" of a metamaterial built from repeated units of this system. We find optical behaviors ranging from that of a simple two-level system (Lorentz-oscillator) to effectively transparent, as the coupling goes from the weak to deep strong coupling regimes. We explore a one-dimensional chain of these meta-atoms, demonstrating the tunability of its optical behavior. Our scheme may ultimately provide a framework for designing new metamaterials with low-loss, highly-confined modes, as well as tunable (single-photon) nonlinearities. △ Less

Submitted 6 September, 2022; originally announced September 2022.

Discovering Conservation Laws using Optimal Transport and Manifold Learning

Authors: Peter Y. Lu, Rumen Dangovski, Marin Soljačić

Abstract: Conservation laws are key theoretical and practical tools for understanding, characterizing, and modeling nonlinear dynamical systems. However, for many complex systems, the corresponding conserved quantities are difficult to identify, making it hard to analyze their dynamics and build stable predictive models. Current approaches for discovering conservation laws often depend on detailed dynamical… ▽ More Conservation laws are key theoretical and practical tools for understanding, characterizing, and modeling nonlinear dynamical systems. However, for many complex systems, the corresponding conserved quantities are difficult to identify, making it hard to analyze their dynamics and build stable predictive models. Current approaches for discovering conservation laws often depend on detailed dynamical information or rely on black box parametric deep learning methods. We instead reformulate this task as a manifold learning problem and propose a non-parametric approach for discovering conserved quantities. We test this new approach on a variety of physical systems and demonstrate that our method is able to both identify the number of conserved quantities and extract their values. Using tools from optimal transport theory and manifold learning, our proposed method provides a direct geometric approach to identifying conservation laws that is both robust and interpretable without requiring an explicit model of the system nor accurate time information. △ Less

Submitted 22 August, 2023; v1 submitted 31 August, 2022; originally announced August 2022.

Comments: 30 pages, 15 figures (7 main text, 8 supplemental), 3 tables (supplemental)

Journal ref: Nat. Commun. 14, 4744 (2023)

Free-electron-light interactions in nanophotonics

Authors: Charles Roques-Carmes, Steven E. Kooi, Yi Yang, Nicholas Rivera, Phillip D. Keathley, John D. Joannopoulos, Steven G. Johnson, Ido Kaminer, Karl K. Berggren, Marin Soljačić

Abstract: When impinging on optical structures or passing in their vicinity, free electrons can spontaneously emit electromagnetic radiation, a phenomenon generally known as cathodoluminescence. Free-electron radiation comes in many guises: Cherenkov, transition, and Smith-Purcell radiation, but also electron scintillation, commonly referred to as incoherent cathodoluminescence. While those effects have bee… ▽ More When impinging on optical structures or passing in their vicinity, free electrons can spontaneously emit electromagnetic radiation, a phenomenon generally known as cathodoluminescence. Free-electron radiation comes in many guises: Cherenkov, transition, and Smith-Purcell radiation, but also electron scintillation, commonly referred to as incoherent cathodoluminescence. While those effects have been at the heart of many fundamental discoveries and technological developments in high-energy physics in the past century, their recent demonstration in photonic and nanophotonic systems has attracted a lot of attention. Those developments arose from predictions that exploit nanophotonics for novel radiation regimes, now becoming accessible thanks to advances in nanofabrication. In general, the proper design of nanophotonic structures can enable shaping, control, and enhancement of free-electron radiation, for any of the above-mentioned effects. Free-electron radiation in nanophotonics opens the way to promising applications, such as widely-tunable integrated light sources from x-ray to THz frequencies, miniaturized particle accelerators, and highly sensitive high-energy particle detectors. Here, we review the emerging field of free-electron radiation in nanophotonics. We first present a general, unified framework to describe free-electron light-matter interaction in arbitrary nanophotonic systems. We then show how this framework sheds light on the physical underpinnings of many methods in the field used to control and enhance free-electron radiation. Namely, the framework points to the central role played by the photonic eigenmodes in controlling the output properties of free-electron radiation (e.g., frequency, directionality, and polarization). [... see full abstract in paper] △ Less

Submitted 17 August, 2022; v1 submitted 3 August, 2022; originally announced August 2022.

Comments: 34 pages; 9 figures

Deep Learning and Symbolic Regression for Discovering Parametric Equations

Authors: Michael Zhang, Samuel Kim, Peter Y. Lu, Marin Soljačić

Abstract: Symbolic regression is a machine learning technique that can learn the governing formulas of data and thus has the potential to transform scientific discovery. However, symbolic regression is still limited in the complexity and dimensionality of the systems that it can analyze. Deep learning on the other hand has transformed machine learning in its ability to analyze extremely complex and high-dim… ▽ More Symbolic regression is a machine learning technique that can learn the governing formulas of data and thus has the potential to transform scientific discovery. However, symbolic regression is still limited in the complexity and dimensionality of the systems that it can analyze. Deep learning on the other hand has transformed machine learning in its ability to analyze extremely complex and high-dimensional datasets. We propose a neural network architecture to extend symbolic regression to parametric systems where some coefficient may vary but the structure of the underlying governing equation remains constant. We demonstrate our method on various analytic expressions, ODEs, and PDEs with varying coefficients and show that it extrapolates well outside of the training domain. The neural network-based architecture can also integrate with other deep learning architectures so that it can analyze high-dimensional data while being trained end-to-end. To this end we integrate our architecture with convolutional neural networks to analyze 1D images of varying spring systems. △ Less

Submitted 28 May, 2023; v1 submitted 1 July, 2022; originally announced July 2022.

Comments: Michael Zhang and Samuel Kim contributed equally to this work. 13 pages, 7 figures

DiffCSE: Difference-based Contrastive Learning for Sentence Embeddings

Authors: Yung-Sung Chuang, Rumen Dangovski, Hongyin Luo, Yang Zhang, Shiyu Chang, Marin Soljačić, Shang-Wen Li, Wen-tau Yih, Yoon Kim, James Glass

Abstract: We propose DiffCSE, an unsupervised contrastive learning framework for learning sentence embeddings. DiffCSE learns sentence embeddings that are sensitive to the difference between the original sentence and an edited sentence, where the edited sentence is obtained by stochastically masking out the original sentence and then sampling from a masked language model. We show that DiffSCE is an instance… ▽ More We propose DiffCSE, an unsupervised contrastive learning framework for learning sentence embeddings. DiffCSE learns sentence embeddings that are sensitive to the difference between the original sentence and an edited sentence, where the edited sentence is obtained by stochastically masking out the original sentence and then sampling from a masked language model. We show that DiffSCE is an instance of equivariant contrastive learning (Dangovski et al., 2021), which generalizes contrastive learning and learns representations that are insensitive to certain types of augmentations and sensitive to other "harmful" types of augmentations. Our experiments show that DiffCSE achieves state-of-the-art results among unsupervised sentence representation learning methods, outperforming unsupervised SimCSE by 2.3 absolute points on semantic textual similarity tasks. △ Less

Submitted 21 April, 2022; originally announced April 2022.

Comments: NAACL 2022 main conference (Long paper). Pretrained models and code are available at https://github.com/voidism/DiffCSE

Topogivity: A Machine-Learned Chemical Rule for Discovering Topological Materials

Authors: Andrew Ma, Yang Zhang, Thomas Christensen, Hoi Chun Po, Li Jing, Liang Fu, Marin Soljačić

Abstract: Topological materials present unconventional electronic properties that make them attractive for both basic science and next-generation technological applications. The majority of currently known topological materials have been discovered using methods that involve symmetry-based analysis of the quantum wavefunction. Here we use machine learning to develop a simple-to-use heuristic chemical rule t… ▽ More Topological materials present unconventional electronic properties that make them attractive for both basic science and next-generation technological applications. The majority of currently known topological materials have been discovered using methods that involve symmetry-based analysis of the quantum wavefunction. Here we use machine learning to develop a simple-to-use heuristic chemical rule that diagnoses with a high accuracy whether a material is topological using only its chemical formula. This heuristic rule is based on a notion that we term topogivity, a machine-learned numerical value for each element that loosely captures its tendency to form topological materials. We next implement a high-throughput procedure for discovering topological materials based on the heuristic topogivity-rule prediction followed by ab initio validation. This way, we discover new topological materials that are not diagnosable using symmetry indicators, including several that may be promising for experimental observation. △ Less

Submitted 23 January, 2023; v1 submitted 10 February, 2022; originally announced February 2022.

Comments: Main text: 6 pages, 3 figures; supplementary materials: 43 pages, 62 figures, 5 tables

Journal ref: Nano Lett. 2023, 23, 3, 772-778

End-to-End Optimization of Metasurfaces for Imaging with Compressed Sensing

Authors: Gaurav Arya, William F. Li, Charles Roques-Carmes, Marin Soljačić, Steven G. Johnson, Zin Lin

Abstract: We present a framework for the end-to-end optimization of metasurface imaging systems that reconstruct targets using compressed sensing, a technique for solving underdetermined imaging problems when the target object exhibits sparsity (i.e. the object can be described by a small number of non-zero values, but the positions of these values are unknown). We nest an iterative, unapproximated compress… ▽ More We present a framework for the end-to-end optimization of metasurface imaging systems that reconstruct targets using compressed sensing, a technique for solving underdetermined imaging problems when the target object exhibits sparsity (i.e. the object can be described by a small number of non-zero values, but the positions of these values are unknown). We nest an iterative, unapproximated compressed sensing reconstruction algorithm into our end-to-end optimization pipeline, resulting in an interpretable, data-efficient method for maximally leveraging metaoptics to exploit object sparsity. We apply our framework to super-resolution imaging and high-resolution depth imaging with a phase-change material. In both situations, our end-to-end framework computationally discovers optimal metasurface structures for compressed sensing recovery, automatically balancing a number of complicated design considerations to select an imaging measurement matrix from a complex, physically constrained manifold with millions ofdimensions. The optimized metasurface imaging systems are robust to noise, significantly improving over random scattering surfaces and approaching the ideal compressed sensing performance of a Gaussian matrix, showing how a physical metasurface system can demonstrably approach the mathematical limits of compressed sensing. △ Less

Submitted 26 June, 2024; v1 submitted 27 January, 2022; originally announced January 2022.

Journal ref: ACS Photonics 2024, 11, 5, 2077-2087

Meta-Learning and Self-Supervised Pretraining for Real World Image Translation

Authors: Ileana Rugina, Rumen Dangovski, Mark Veillette, Pooya Khorrami, Brian Cheung, Olga Simek, Marin Soljačić

Abstract: Recent advances in deep learning, in particular enabled by hardware advances and big data, have provided impressive results across a wide range of computational problems such as computer vision, natural language, or reinforcement learning. Many of these improvements are however constrained to problems with large-scale curated data-sets which require a lot of human labor to gather. Additionally, th… ▽ More Recent advances in deep learning, in particular enabled by hardware advances and big data, have provided impressive results across a wide range of computational problems such as computer vision, natural language, or reinforcement learning. Many of these improvements are however constrained to problems with large-scale curated data-sets which require a lot of human labor to gather. Additionally, these models tend to generalize poorly under both slight distributional shifts and low-data regimes. In recent years, emerging fields such as meta-learning or self-supervised learning have been closing the gap between proof-of-concept results and real-life applications of machine learning by extending deep-learning to the semi-supervised and few-shot domains. We follow this line of work and explore spatio-temporal structure in a recently introduced image-to-image translation problem in order to: i) formulate a novel multi-task few-shot image generation benchmark and ii) explore data augmentations in contrastive pre-training for image translation downstream tasks. We present several baselines for the few-shot problem and discuss trade-offs between different approaches. Our code is available at https://github.com/irugina/meta-image-translation. △ Less

Submitted 22 December, 2021; originally announced December 2021.

Comments: 10 pages, 8 figures, 2 tables

Nonperturbative electromagnetic nonlinearities, n-photon reflectors, and Fock-state lasers based on deep-strong coupling of light and matter

Authors: Nicholas Rivera, Jamison Sloan, Ido Kaminer, Marin Soljacic

Abstract: Light and matter can now interact in a regime where their coupling is stronger than their bare energies. This deep-strong coupling (DSC) regime of quantum electrodynamics promises to challenge many conventional assumptions about the physics of light and matter. Here, we show how light and matter interactions in this regime give rise to electromagnetic nonlinearities dramatically different from tho… ▽ More Light and matter can now interact in a regime where their coupling is stronger than their bare energies. This deep-strong coupling (DSC) regime of quantum electrodynamics promises to challenge many conventional assumptions about the physics of light and matter. Here, we show how light and matter interactions in this regime give rise to electromagnetic nonlinearities dramatically different from those of naturally existing materials. Excitations in the DSC regime act as photons with a linear energy spectrum up to a critical excitation number, after which, the system suddenly becomes strongly anharmonic, thus acting as an effective intensity-dependent nonlinearity of an extremely high order. We show that this behavior allows for N-photon blockade (with $N \gg 1$), enabling qualitatively new kinds of quantum light sources. For example, this nonlinearity forms the basis for a new type of gain medium, which when integrated into a laser or maser, produces large Fock states (rather than coherent states). Such Fock states could in principle have photon numbers orders of magnitude larger than any realized previously, and would be protected from dissipation by a new type of equilibrium between nonlinear gain and linear loss. We discuss paths to experimental realization of the effects described here. △ Less

Submitted 26 May, 2023; v1 submitted 12 November, 2021; originally announced November 2021.

Complete condensation of photon noise in nonlinear dissipative systems

Authors: Nicholas Rivera, Jamison Sloan, Yannick Salamin, Marin Soljacic

Abstract: Fock states are the most fundamental quantum states of bosonic fields, forming an important basis for understanding their quantum dynamics. As energy and number eigenstates, they have an exactly defined number of quanta, and most faithfully express the particle nature of fields. These properties make them attractive for many applications in metrology, and quantum simulation and information process… ▽ More Fock states are the most fundamental quantum states of bosonic fields, forming an important basis for understanding their quantum dynamics. As energy and number eigenstates, they have an exactly defined number of quanta, and most faithfully express the particle nature of fields. These properties make them attractive for many applications in metrology, and quantum simulation and information processing. Yet, Fock states are notoriously difficult to generate. The problem is especially acute in optics, where it is difficult to deterministically produce Fock states with more than a single photon, let alone at macroscopic scales. This is in part due to a dearth of mechanisms to produce large Fock states, as well as the deleterious effects of linear dissipation. Here, we introduce a new effect in the physics of nonlinear bosons, arising from the interplay of dissipation and Kerr nonlinearity. In this effect, a nonlinear resonance is dissipationless when it has a particular number of quanta (e.g., photons) inside it, and lossy otherwise. This loss, which results from nonlinear interference, leads to several new quantum statistical effects. For example, it leads to spontaneous condensation of intensity noise, which may enable generation of large Fock and extremely photon-number-squeezed states of light. We also show how this effect has implications for new classes of optoelectronic devices such as lasers, which can stabilize extremely low-noise states in an equilibrium between gain and the nonlinear loss that we introduce. Throughout the text, we present examples of systems that may realize these effects. In one, we show how the nonlinear dissipation could lead to optical Fock states of $n=1000$, while in another, we show how conventional laser architectures could be used to generate macroscopic light ($>10^{12}$ photons) with nearly 95% less noise than the standard quantum limit. △ Less

Submitted 20 February, 2022; v1 submitted 4 November, 2021; originally announced November 2021.

End-to-end metasurface inverse design for single-shot multi-channel imaging

Authors: Zin Lin, Raphaël Pestourie, Charles Roques-Carmes, Zhaoyi Li, Federico Capasso, Marin Soljačić, Steven G. Johnson

Abstract: We introduce end-to-end metaoptics inverse design for multi-channel imaging: reconstruction of depth, spectral and polarization channels from a single-shot monochrome image. The proposed technique integrates a single-layer metasurface frontend with an efficient Tikhonov reconstruction backend, without any additional optics except a grayscale sensor. Our method yields multi-channel imaging by spont… ▽ More We introduce end-to-end metaoptics inverse design for multi-channel imaging: reconstruction of depth, spectral and polarization channels from a single-shot monochrome image. The proposed technique integrates a single-layer metasurface frontend with an efficient Tikhonov reconstruction backend, without any additional optics except a grayscale sensor. Our method yields multi-channel imaging by spontaneous demultiplexing: the metaoptics front-end separates different channels into distinct spatial domains whose locations on the sensor are optimally discovered by the inverse-design algorithm. We present large-area metasurface designs, compatible with standard lithography, for multi-spectral imaging, depth-spectral imaging, and ``all-in-one'' spectro-polarimetric-depth imaging with robust reconstruction performance ($\lesssim 10\%$ error with 1\% detector noise). In contrast to neural networks, our framework is physically interpretable and does not require large training sets. It can be used to reconstruct arbitrary three-dimensional scenes with full multi-wavelength spectra and polarization textures. △ Less

Submitted 1 November, 2021; originally announced November 2021.

Equivariant Contrastive Learning

Authors: Rumen Dangovski, Li Jing, Charlotte Loh, Seungwook Han, Akash Srivastava, Brian Cheung, Pulkit Agrawal, Marin Soljačić

Abstract: In state-of-the-art self-supervised learning (SSL) pre-training produces semantically good representations by encouraging them to be invariant under meaningful transformations prescribed from human knowledge. In fact, the property of invariance is a trivial instance of a broader class called equivariance, which can be intuitively understood as the property that representations transform according… ▽ More In state-of-the-art self-supervised learning (SSL) pre-training produces semantically good representations by encouraging them to be invariant under meaningful transformations prescribed from human knowledge. In fact, the property of invariance is a trivial instance of a broader class called equivariance, which can be intuitively understood as the property that representations transform according to the way the inputs transform. Here, we show that rather than using only invariance, pre-training that encourages non-trivial equivariance to some transformations, while maintaining invariance to other transformations, can be used to improve the semantic quality of representations. Specifically, we extend popular SSL methods to a more general framework which we name Equivariant Self-Supervised Learning (E-SSL). In E-SSL, a simple additional pre-training objective encourages equivariance by predicting the transformations applied to the input. We demonstrate E-SSL's effectiveness empirically on several popular computer vision benchmarks, e.g. improving SimCLR to 72.5% linear probe accuracy on ImageNet. Furthermore, we demonstrate usefulness of E-SSL for applications beyond computer vision; in particular, we show its utility on regression problems in photonics science. Our code, datasets and pre-trained models are available at https://github.com/rdangovs/essl to aid further research in E-SSL. △ Less

Submitted 14 March, 2022; v1 submitted 28 October, 2021; originally announced November 2021.

Comments: Camera Ready Revision. ICLR 2022. Discussion: https://openreview.net/forum?id=gKLAAfiytI Code: https://github.com/rdangovs/essl

A general framework for scintillation in nanophotonics

Authors: Charles Roques-Carmes, Nicholas Rivera, Ali Ghorashi, Steven E. Kooi, Yi Yang, Zin Lin, Justin Beroz, Aviram Massuda, Jamison Sloan, Nicolas Romeo, Yang Yu, John D. Joannopoulos, Ido Kaminer, Steven G. Johnson, Marin Soljačić

Abstract: Bombardment of materials by high-energy particles (e.g., electrons, nuclei, X- and $γ$-ray photons) often leads to light emission, known generally as scintillation. Scintillation is ubiquitous and enjoys widespread applications in many areas such as medical imaging, X-ray non-destructive inspection, night vision, electron microscopy, and high-energy particle detectors. A large body of research foc… ▽ More Bombardment of materials by high-energy particles (e.g., electrons, nuclei, X- and $γ$-ray photons) often leads to light emission, known generally as scintillation. Scintillation is ubiquitous and enjoys widespread applications in many areas such as medical imaging, X-ray non-destructive inspection, night vision, electron microscopy, and high-energy particle detectors. A large body of research focuses on finding new materials optimized for brighter, faster, and more controlled scintillation. Here, we develop a fundamentally different approach based on integrating nanophotonic structures into scintillators to enhance their emission. To start, we develop a unified and ab initio theory of nanophotonic scintillators that accounts for the key aspects of scintillation: the energy loss by high-energy particles, as well as the light emission by non-equilibrium electrons in arbitrary nanostructured optical systems. This theoretical framework allows us, for the first time, to experimentally demonstrate nearly an order-of-magnitude enhancement of scintillation, in both electron-induced, and X-ray-induced scintillation. Our theory also allows the discovery of structures that could eventually achieve several orders-of-magnitude scintillation enhancement. The framework and results shown here should enable the development of a new class of brighter, faster, and higher-resolution scintillators with tailored and optimized performances - with many potential applications where scintillators are used. △ Less

Submitted 21 October, 2021; originally announced October 2021.

Surrogate- and invariance-boosted contrastive learning for data-scarce applications in science

Authors: Charlotte Loh, Thomas Christensen, Rumen Dangovski, Samuel Kim, Marin Soljacic

Abstract: Deep learning techniques have been increasingly applied to the natural sciences, e.g., for property prediction and optimization or material discovery. A fundamental ingredient of such approaches is the vast quantity of labelled data needed to train the model; this poses severe challenges in data-scarce settings where obtaining labels requires substantial computational or labor resources. Here, we… ▽ More Deep learning techniques have been increasingly applied to the natural sciences, e.g., for property prediction and optimization or material discovery. A fundamental ingredient of such approaches is the vast quantity of labelled data needed to train the model; this poses severe challenges in data-scarce settings where obtaining labels requires substantial computational or labor resources. Here, we introduce surrogate- and invariance-boosted contrastive learning (SIB-CL), a deep learning framework which incorporates three ``inexpensive'' and easily obtainable auxiliary information sources to overcome data scarcity. Specifically, these are: 1)~abundant unlabeled data, 2)~prior knowledge of symmetries or invariances and 3)~surrogate data obtained at near-zero cost. We demonstrate SIB-CL's effectiveness and generality on various scientific problems, e.g., predicting the density-of-states of 2D photonic crystals and solving the 3D time-independent Schrodinger equation. SIB-CL consistently results in orders of magnitude reduction in the number of labels needed to achieve the same network accuracies. △ Less

Submitted 15 October, 2021; originally announced October 2021.

Comments: 21 pages, 10 figures

Observation of enhanced free-electron radiation from photonic flatband resonances

Authors: Yi Yang, Charles Roques-Carmes, Steven E. Kooi, Haoning Tang, Justin Beroz, Eric Mazur, Ido Kaminer, John D. Joannopoulos, Marin Soljačić

Abstract: Flatbands emerge from a myriad of structures such as Landau levels, Lieb and Kagome lattices, linegraphs, and more recently moire superlattices. They enable unique properties including slow light in photonics, correlated phases in electronics, and supercollimation in both systems. Despite these intense parallel efforts, flatbands have never been shown to affect the core light-matter interaction be… ▽ More Flatbands emerge from a myriad of structures such as Landau levels, Lieb and Kagome lattices, linegraphs, and more recently moire superlattices. They enable unique properties including slow light in photonics, correlated phases in electronics, and supercollimation in both systems. Despite these intense parallel efforts, flatbands have never been shown to affect the core light-matter interaction between electrons and photons, which is limited by a dimensionality mismatch. Here, we reveal that a photonic flatband can overcome this mismatch between localized electrons and extended photons and thus remarkably boost their light-matter interaction. We design flatband resonances in a silicon-on-insulator photonic crystal slab to control and enhance the radiation emission from free electrons by tuning their trajectory and velocity. In particular, we record a 100-fold radiation enhancement from the conventional diffraction-enabled Smith-Purcell radiation, and show the potential of our approach to achieve $10^6$-fold enhancements and beyond. The enhancement also enables us to perform polarization shaping of free electron radiation from multiple flatbands and demonstrate an approach to measure photonic bands via angle-resolved electron-beam measurements. Our results suggest flatbands as ideal test beds for strong light-electron interaction in various systems, with particular relevance for efficient and compact free-electron light sources and accelerators. △ Less

Submitted 7 October, 2021; originally announced October 2021.

Comments: 7 pages, 4 figures, supplementary included as ancillary file

Journal ref: Nature 613, 42 (2023)

A Brewster route to Cherenkov detectors

Authors: Xiao Lin, Hao Hu, Sajan Easo, Yi Yang, Yichen Shen, Kezhen Yin, Michele Piero Blago, Ido Kaminer, Baile Zhang, Hongsheng Chen, John Joannopoulos, Marin Soljačić, Yu Luo

Abstract: The Cherenkov effect enables a valuable tool, known as the Cherenkov detector, to identify high-energy particles via the measurement of the Cherenkov cone. However, the sensitivity and momentum coverage of such detectors are intrinsically limited by the refractive index of the host material. Especially, identifying particles with energy above multiple gigaelectronvolts requires host materials with… ▽ More The Cherenkov effect enables a valuable tool, known as the Cherenkov detector, to identify high-energy particles via the measurement of the Cherenkov cone. However, the sensitivity and momentum coverage of such detectors are intrinsically limited by the refractive index of the host material. Especially, identifying particles with energy above multiple gigaelectronvolts requires host materials with a near-unity refractive index, which are often limited to large and bulky gas chambers. Overcoming this fundamental material limit is important for future particle detectors yet remains a long-standing scientific challenge. Here, we propose a different paradigm for Cherenkov detectors that utilizes the broadband angular filter made from stacks of variable one-dimensional photonic crystals. Owing to the Brewster effect, the angular filter is transparent only to Cherenkov photons from a precise incident angle, and particle identification is achieved by mapping each Cherenkov angle to the peak-intensity position of transmitted photons in the detection plane. This unique property of the angular filter is exceptionally beneficial to Cherenkov detection as it enables the realization of a non-dispersive pseudo refractive index over the entire visible spectrum. Moreover, such a pseudo refractive index can be flexibly tuned to arbitrary values, including those close to unity. Our angular-selective Brewster paradigm offers a feasible solution to implement compact and highly sensitive Cherenkov detectors especially in beam lines and it can cover a wide momentum range using readily available dielectric materials. △ Less

Submitted 26 July, 2021; originally announced July 2021.