-
Automated and Holistic Co-design of Neural Networks and ASICs for Enabling In-Pixel Intelligence
Authors:
Shubha R. Kharel,
Prashansa Mukim,
Piotr Maj,
Grzegorz W. Deptuch,
Shinjae Yoo,
Yihui Ren,
Soumyajit Mandal
Abstract:
Extreme edge-AI systems, such as those in readout ASICs for radiation detection, must operate under stringent hardware constraints such as micron-level dimensions, sub-milliwatt power, and nanosecond-scale speed while providing clear accuracy advantages over traditional architectures. Finding ideal solutions means identifying optimal AI and ASIC design choices from a design space that has explosiv…
▽ More
Extreme edge-AI systems, such as those in readout ASICs for radiation detection, must operate under stringent hardware constraints such as micron-level dimensions, sub-milliwatt power, and nanosecond-scale speed while providing clear accuracy advantages over traditional architectures. Finding ideal solutions means identifying optimal AI and ASIC design choices from a design space that has explosively expanded during the merger of these domains, creating non-trivial couplings which together act upon a small set of solutions as constraints tighten. It is impractical, if not impossible, to manually determine ideal choices among possibilities that easily exceed billions even in small-size problems. Existing methods to bridge this gap have leveraged theoretical understanding of hardware to f architecture search. However, the assumptions made in computing such theoretical metrics are too idealized to provide sufficient guidance during the difficult search for a practical implementation. Meanwhile, theoretical estimates for many other crucial metrics (like delay) do not even exist and are similarly variable, dependent on parameters of the process design kit (PDK). To address these challenges, we present a study that employs intelligent search using multi-objective Bayesian optimization, integrating both neural network search and ASIC synthesis in the loop. This approach provides reliable feedback on the collective impact of all cross-domain design choices. We showcase the effectiveness of our approach by finding several Pareto-optimal design choices for effective and efficient neural networks that perform real-time feature extraction from input pulses within the individual pixels of a readout ASIC.
△ Less
Submitted 18 July, 2024;
originally announced July 2024.
-
The sequence of prime gaps is graphic
Authors:
Péter L. Erdős,
Gergely Harcos,
Shubha R. Kharel,
Péter Maga,
Tamás R. Mezei,
Zoltán Toroczkai
Abstract:
Let us call a simple graph on $n\geq 2$ vertices a prime gap graph if its vertex degrees are $1$ and the first $n-1$ prime gaps. We show that such a graph exists for every large $n$, and in fact for every $n\geq 2$ if we assume the Riemann hypothesis. Moreover, an infinite sequence of prime gap graphs can be generated by the so-called degree preserving growth process. This is the first time a natu…
▽ More
Let us call a simple graph on $n\geq 2$ vertices a prime gap graph if its vertex degrees are $1$ and the first $n-1$ prime gaps. We show that such a graph exists for every large $n$, and in fact for every $n\geq 2$ if we assume the Riemann hypothesis. Moreover, an infinite sequence of prime gap graphs can be generated by the so-called degree preserving growth process. This is the first time a naturally occurring infinite sequence of positive integers is identified as graphic. That is, we show the existence of an interesting, and so far unique, infinite combinatorial object.
△ Less
Submitted 10 January, 2023; v1 submitted 1 May, 2022;
originally announced May 2022.
-
On graphic degree sequences and matching numbers
Authors:
Péter L. Erdős,
Shubha R. Kharel,
Tamás Róbert Mezei,
Zoltán Toroczkai
Abstract:
The recently introduced Degree Preserving Growth model (Nature Physics, DOI:10.1038/s41567-021-01417-7) uses matchings to insert new vertices of prescribed degrees into the current graph of an ever-growing graph sequence. In this paper we are presenting lower bounds on the matching numbers of the graphs in the growth sequence, based on their degree sequence only. We are interested in bounds on max…
▽ More
The recently introduced Degree Preserving Growth model (Nature Physics, DOI:10.1038/s41567-021-01417-7) uses matchings to insert new vertices of prescribed degrees into the current graph of an ever-growing graph sequence. In this paper we are presenting lower bounds on the matching numbers of the graphs in the growth sequence, based on their degree sequence only. We are interested in bounds on maximal/maximum matchings in some realizations (potential matching number), and also on common bounds on all realizations (forcible matching number).
Particularly, we prove that whether a degree sequence can be extended with a new degree $2δ$ is equivalent to the existence of a realization of the original degree sequence with a matching of size $δ$ that covers the largest $2δ$ degrees. Second, we estimate the minimum size of both the maximal and the maximum matchings, as determined by the degree sequence, independently of the graphical realization. Furthermore, we also estimate the maximum value of the matching number over all possible realizations for the degree sequence. Along this line we answer a question raised by Biedl, Demaine et. al. (DOI:10.1016/j.disc.2004.05.003). We briefly discuss an application to prime gap sequences.
△ Less
Submitted 4 January, 2024; v1 submitted 15 April, 2022;
originally announced April 2022.
-
Degree-preserving graph dynamics -- a versatile process to construct random networks
Authors:
Péter L. Erdős,
Shubha R. Kharel,
Tamás R. Mezei,
Zoltán Toroczkai
Abstract:
Real-world networks evolve over time via additions or removals of vertices and edges. In current network evolution models, vertex degree varies or grows arbitrarily. A recently introduced degree-preserving network growth (DPG) family of models preserves vertex degree, resulting in structures significantly different from and more diverse than previous models ([Nature Physics 2021, DOI: 10.1038/s415…
▽ More
Real-world networks evolve over time via additions or removals of vertices and edges. In current network evolution models, vertex degree varies or grows arbitrarily. A recently introduced degree-preserving network growth (DPG) family of models preserves vertex degree, resulting in structures significantly different from and more diverse than previous models ([Nature Physics 2021, DOI: 10.1038/s41567-021-01417-7]). Despite its degree preserving property, the DPG model is able to replicate the output of several well-known real-world network growth models. Simulations showed that many well-studied real-world networks can be constructed from small seed graphs.
Here we start the development of a rigorous mathematical theory underlying the DPG family of network growth models. We prove that the degree sequence of the output of some of the well-known, real-world network growth models can be reconstructed via the DPG process, using proper parametrization. We also show that the general problem of deciding whether a simple graph can be obtained via the DPG process from a small seed (DPG feasibility) is, as expected, NP-complete. It is an important open problem to uncover whether there is a structural reason behind the DPG-constructibility of real-world networks.
△ Less
Submitted 12 October, 2022; v1 submitted 23 November, 2021;
originally announced November 2021.
