-
Modelling multiplex testing for outbreak Control
Authors:
Martyn Fyles,
Christopher E. Overton,
Tom Ward,
Emma Bennett,
Tom Fowler,
Ian Hall
Abstract:
During the SARS-CoV-2 pandemic, polymerase chain reaction (PCR) and lateral flow device (LFD) tests were frequently deployed to detect the presence of SARS-CoV-2. Many of these tests were singleplex, and only tested for the presence of a single pathogen. Multiplex tests can test for the presence of several pathogens using only a single swab, which can allow for: surveillance of more pathogens, tar…
▽ More
During the SARS-CoV-2 pandemic, polymerase chain reaction (PCR) and lateral flow device (LFD) tests were frequently deployed to detect the presence of SARS-CoV-2. Many of these tests were singleplex, and only tested for the presence of a single pathogen. Multiplex tests can test for the presence of several pathogens using only a single swab, which can allow for: surveillance of more pathogens, targeting of antiviral interventions, a reduced burden of testing, and lower costs. Test sensitivity however, particularly in LFD tests, is highly conditional on the viral concentration dynamics of individuals. To inform the use of multiplex testing in outbreak detection it is therefore necessary to investigate the interactions between outbreak detection strategies and the differing viral concentration trajectories of key pathogens. Viral concentration trajectories are estimated for SARS-CoV-2, and Influenza A/B. Testing strategies for the first five symptomatic cases in an outbreak are then simulated and used to evaluate key performance indicators. Strategies that use a combination of multiplex LFD and PCR tests achieve; high levels of detection, detect outbreaks rapidly, and have the lowest burden of testing across multiple pathogens. Influenza B was estimated to have lower rates of detection due to its modelled viral concentration dynamics.
△ Less
Submitted 30 August, 2024;
originally announced August 2024.
-
The Llama 3 Herd of Models
Authors:
Abhimanyu Dubey,
Abhinav Jauhri,
Abhinav Pandey,
Abhishek Kadian,
Ahmad Al-Dahle,
Aiesha Letman,
Akhil Mathur,
Alan Schelten,
Amy Yang,
Angela Fan,
Anirudh Goyal,
Anthony Hartshorn,
Aobo Yang,
Archi Mitra,
Archie Sravankumar,
Artem Korenev,
Arthur Hinsvark,
Arun Rao,
Aston Zhang,
Aurelien Rodriguez,
Austen Gregerson,
Ava Spataru,
Baptiste Roziere,
Bethany Biron,
Binh Tang
, et al. (510 additional authors not shown)
Abstract:
Modern artificial intelligence (AI) systems are powered by foundation models. This paper presents a new set of foundation models, called Llama 3. It is a herd of language models that natively support multilinguality, coding, reasoning, and tool usage. Our largest model is a dense Transformer with 405B parameters and a context window of up to 128K tokens. This paper presents an extensive empirical…
▽ More
Modern artificial intelligence (AI) systems are powered by foundation models. This paper presents a new set of foundation models, called Llama 3. It is a herd of language models that natively support multilinguality, coding, reasoning, and tool usage. Our largest model is a dense Transformer with 405B parameters and a context window of up to 128K tokens. This paper presents an extensive empirical evaluation of Llama 3. We find that Llama 3 delivers comparable quality to leading language models such as GPT-4 on a plethora of tasks. We publicly release Llama 3, including pre-trained and post-trained versions of the 405B parameter language model and our Llama Guard 3 model for input and output safety. The paper also presents the results of experiments in which we integrate image, video, and speech capabilities into Llama 3 via a compositional approach. We observe this approach performs competitively with the state-of-the-art on image, video, and speech recognition tasks. The resulting models are not yet being broadly released as they are still under development.
△ Less
Submitted 15 August, 2024; v1 submitted 31 July, 2024;
originally announced July 2024.
-
Accreting Primordial Black Holes: Dark Matter Constituents
Authors:
Brandon Curd,
Richard Anantua,
T. Kenneth Fowler
Abstract:
This paper shows that accretion of positronium plasma between 0.01 to 14s after the Big Bang could have created small black holes contributing at least 1 percent of the dark matter present today, with uncertainties ranging from 10 percent or more. General relativistic magnetohydrodynamic (GRMHD) simulations newly adapted to the early Universe confirm that accretion is due to magneto-rotational ins…
▽ More
This paper shows that accretion of positronium plasma between 0.01 to 14s after the Big Bang could have created small black holes contributing at least 1 percent of the dark matter present today, with uncertainties ranging from 10 percent or more. General relativistic magnetohydrodynamic (GRMHD) simulations newly adapted to the early Universe confirm that accretion is due to magneto-rotational instability (MRI) in a rotating plasma. By contrast with Bondi accretion producing primordial masses bigger than the Sun, MRI could produce masses 10^{15-18} g observable by their Hawking radiation contributing to background gamma rays.
△ Less
Submitted 7 March, 2024;
originally announced March 2024.
-
Accreting Primordial Black Holes as Dark Matter Constituents
Authors:
T. Kenneth Fowler,
Richard Anantua
Abstract:
We show how magnetic accretion of positronium (electron-positron) plasma by primordial black holes might significantly contribute to the mass of dark matter in the present Universe. Assuming that background gamma radiation is primordial black hole Hawking radiation rules out Bondi accretion, while magnetic accretion known from studies of active galactic nuclei could explain the abundance of dark m…
▽ More
We show how magnetic accretion of positronium (electron-positron) plasma by primordial black holes might significantly contribute to the mass of dark matter in the present Universe. Assuming that background gamma radiation is primordial black hole Hawking radiation rules out Bondi accretion, while magnetic accretion known from studies of active galactic nuclei could explain the abundance of dark matter. Various accretion scenarios are discussed.
△ Less
Submitted 4 March, 2024; v1 submitted 16 March, 2023;
originally announced March 2023.
-
Optical phased array neural probes for beam-steering in brain tissue
Authors:
Wesley D. Sacher,
Fu-Der Chen,
Homeira Moradi-Chameh,
Xinyu Liu,
Ilan Felts Almog,
Thomas Lordello,
Michael Chang,
Azadeh Naderian,
Trevor M. Fowler,
Eran Segev,
Tianyuan Xue,
Sara Mahallati,
Taufik A. Valiante,
Laurent C. Moreaux,
Joyce K. S. Poon,
Michael L. Roukes
Abstract:
Implantable silicon neural probes with integrated nanophotonic waveguides can deliver patterned dynamic illumination into brain tissue at depth. Here, we introduce neural probes with integrated optical phased arrays and demonstrate optical beam steering in vitro. Beam formation in brain tissue was simulated and characterized. The probes were used for optogenetic stimulation and calcium imaging.
Implantable silicon neural probes with integrated nanophotonic waveguides can deliver patterned dynamic illumination into brain tissue at depth. Here, we introduce neural probes with integrated optical phased arrays and demonstrate optical beam steering in vitro. Beam formation in brain tissue was simulated and characterized. The probes were used for optogenetic stimulation and calcium imaging.
△ Less
Submitted 10 August, 2021;
originally announced August 2021.
-
Hyper-Resistive Model of Ultra High Energy Cosmic Ray Acceleration by Magnetically Collimated Jets Created by Active Galactic Nuclei
Authors:
T. Kenneth Fowler,
Hui Li,
Richard Anantua
Abstract:
This is the fourth in a series of companion papers showing that, when an efficient dynamo can be maintained by accretion disks around supermassive black holes in Active Galactic Nuclei (AGNs), it will lead to the formation of a powerful, magnetically-collimated helix that could explain both the observed jet/radiolobe structures on very large scales and ultimately the enormous power inferred from t…
▽ More
This is the fourth in a series of companion papers showing that, when an efficient dynamo can be maintained by accretion disks around supermassive black holes in Active Galactic Nuclei (AGNs), it will lead to the formation of a powerful, magnetically-collimated helix that could explain both the observed jet/radiolobe structures on very large scales and ultimately the enormous power inferred from the observed ultra high energy cosmic rays (UHECRs) with energies > 10^19 eV. Many timescales are involved in this process. Our hyper-resistive magnetohydrodynamic (MHD) model provides a bridge between General Relativistic MHD simulations of dynamo formation, on the short accretion timescale, and observational evidence of magnetic collimation of large-scale jets on astrophysical timescales. Given the final magnetic structure, we apply hyper-resistive kinetic theory to show how instability causes slowly-evolving magnetically-collimated jets to become the most powerful relativistic accelerators in the Universe. The model yields nine observables in reasonable agreement with observations: the jet length, radiolobe radius and apparent opening angle as observed by synchrotron radiation; the synchrotron total power, synchrotron wavelengths and maximum electron energy (TeVs); and the maximum UHECR energy, the cosmic ray energy spectrum and the cosmic ray intensity on Earth.
△ Less
Submitted 13 September, 2019; v1 submitted 15 March, 2019;
originally announced March 2019.
-
Managing uncertainty in data-derived densities to accelerate density functional theory
Authors:
Andrew T. Fowler,
Chris J. Pickard,
James A. Elliott
Abstract:
Faithful representations of atomic environments and general models for regression can be harnessed to learn electron densities that are close to the ground state. One of the applications of data-derived electron densities is to orbital-free density functional theory. However, extrapolations of densities learned from a training set to dissimilar structures could result in inaccurate results, which…
▽ More
Faithful representations of atomic environments and general models for regression can be harnessed to learn electron densities that are close to the ground state. One of the applications of data-derived electron densities is to orbital-free density functional theory. However, extrapolations of densities learned from a training set to dissimilar structures could result in inaccurate results, which would limit the applicability of the method. Here, we show that a non-Bayesian approach can produce estimates of uncertainty which can successfully distinguish accurate from inaccurate predictions of electron density. We apply our approach to density functional theory where we initialise calculations with data-derived densities only when we are confident about their quality. This results in a guaranteed acceleration to self-consistency for configurations that are similar to those seen during training and could be useful for sampling based methods, where previous ground state densities cannot be used to initialise subsequent calculations.
△ Less
Submitted 28 February, 2019; v1 submitted 5 December, 2018;
originally announced December 2018.
-
Injection and extraction magnets: kicker magnets
Authors:
M. J. Barnes,
L. Ducimetiére,
T. Fowler,
V. Senaj,
L. Sermeus
Abstract:
Each stage of an accelerator system has a limited dynamic range and therefore a chain of stages is required to reach high energy. A combination of septa and kicker magnets is frequently used to inject and extract beam from each stage. The kicker magnets typically produce rectangular field pulses with fast rise- and/or fall-times, however, the field strength is relatively low. To compensate for the…
▽ More
Each stage of an accelerator system has a limited dynamic range and therefore a chain of stages is required to reach high energy. A combination of septa and kicker magnets is frequently used to inject and extract beam from each stage. The kicker magnets typically produce rectangular field pulses with fast rise- and/or fall-times, however, the field strength is relatively low. To compensate for their relatively low field strength, the kicker magnets are generally combined with electromagnetic septa. The septa provide relatively strong field strength but are either DC or slow pulsed. This paper discusses injection and extraction systems with particular emphasis on the hardware required for the kicker magnet.
△ Less
Submitted 8 March, 2011;
originally announced March 2011.
-
Finding Large Monochromatic Diameter Two Subgraphs
Authors:
Tom Fowler
Abstract:
Given a coloring of the edges of the complete graph on n vertices in k colors, by considering the neighbors of an arbitrary vertex it follows that there is a monochromatic diameter two subgraph on at least 1+(n-1)/k vertices. We show that for $k \ge 3$ this is asymptotically best possible, and that for k=2 there is always a monochromatic diameter two subgraph on at least…
▽ More
Given a coloring of the edges of the complete graph on n vertices in k colors, by considering the neighbors of an arbitrary vertex it follows that there is a monochromatic diameter two subgraph on at least 1+(n-1)/k vertices. We show that for $k \ge 3$ this is asymptotically best possible, and that for k=2 there is always a monochromatic diameter two subgraph on at least $\lceil {3 \over 4} n \rceil$ vertices, which again, is best possible.
△ Less
Submitted 30 August, 1999;
originally announced August 1999.
