arXiv subjects

Jonathan Thomas

Publications and source records attributed to Jonathan Thomas.

6 recordsLinked to original sources

arXiv2026

Full-Self Diagnostics (FSD): Physics-Grounded Visual Biomarker Inference from Smartphone Video via Inverse Problems and Operator Learning

We present Full-Self Diagnostics (FSD), a unified mathematical framework for recovering latent physiological states from unconstrained 9-second facial videos captured by consumer smartphones. The approach integrates five mutually reinforcing components: (1) a physics-based forward model derived from the radiative transfer equation and chromophore absorption that maps camera observables to biomarker concentrations; (2) an information-theoretic observability theory proving that multi-channel visual signals (spectral, pulse, respiratory, micro-expression, and oculomotor) contain strictly increasing mutual information with physiological state; (3) a stable, Tikhonov-regularized inverse problem with domain-uniform identifiability guarantees; (4) an operator-learning formulation that enables generalization across devices, resolutions, and populations; and (5) a supervised learning procedure, interpretable as stochastic variational inference, that continuously refines the model from paired biosensor ground truth with performance improving proportionally to one over the square root of the number of paired observations. Empirical validation on 38812 real-world paired scans across 59 subjects demonstrates practical performance. Self-collected data from the lead author (glucose range 35-550 mg/dL) yields MARD of 29.86 percent with 97.57 percent of predictions in Clarke Error Grid Zones A+B and only 0.27 percent in the dangerous Zone E. A well-managed diabetic participant achieves MARD of 17 percent in the narrower 70-180 mg/dL band. These results confirm that consumer-grade facial video encodes sufficient structured information for clinically relevant, non-invasive biomarker inference under fully unconstrained conditions, with performance scaling predictably as more paired data becomes available.

eess.IV↗

arXiv2024

Learning Efficient Recursive Numeral Systems via Reinforcement Learning

It has previously been shown that by using reinforcement learning (RL), agents can derive simple approximate and exact-restricted numeral systems that are similar to human ones (Carlsson, 2021). However, it is a major challenge to show how more complex recursive numeral systems, similar to for example English, could arise via a simple learning mechanism such as RL. Here, we introduce an approach towards deriving a mechanistic explanation of the emergence of efficient recursive number systems. We consider pairs of agents learning how to communicate about numerical quantities through a meta-grammar that can be gradually modified throughout the interactions. Utilising a slightly modified version of the meta-grammar of Hurford (1975), we demonstrate that our RL agents, shaped by the pressures for efficient communication, can effectively modify their lexicon towards Pareto-optimal configurations which are comparable to those observed within human numeral systems in terms of their efficiency.

cs.CL↗

arXiv2023

Kinetic relaxation and nucleation of Bose stars in self-interacting wave dark matter

We revisit kinetic relaxation and soliton/Boson star nucleation in fuzzy scalar dark matter featuring short-ranged self-interactions $\mathcal{H}_{\rm int} = -\lambda|\psi|^4/2m^2$, alongside gravitational self-interactions. We map out the full curve of nucleation timescale for both repulsive ($\lambda < 0$) and attractive ($\lambda > 0$) short-ranged self-interaction strength, and in doing so reveal two new points. Firstly, besides the two usual terms, $\propto G^2$ and $\propto \lambda^2$, in the total relaxation rate $\Gamma_{\rm relax}$, there is an additional cross term $\propto G\lambda$ arising due to interference between gravitational and short-ranged self-interaction scattering amplitudes. This yields a critical repulsive interaction strength $\lambda_{\rm cr} \simeq - 2\pi Gm^2/v_{0}^2$, at which the relaxation rate is smallest and serves as the transition point between typical net attractive self-interaction ($\lambda \gtrsim \lambda_{\rm cr}$), and net repulsive self-interaction ($-\lambda \gtrsim -\lambda_{\rm cr}$). Secondly, while in the net attractive regime, nucleation time scale is similar to inverse relaxation time scale $\tau_{\rm nuc} \sim \Gamma^{-1}_{\rm relax}$, in the net repulsive regime nucleation occurs at a delayed time $\tau_{\rm nuc} \sim (\lambda/\lambda_{\rm cr})\Gamma^{-1}_{\rm relax}$. We confirm our analytical understanding by performing 3D field simulations with varying average mass density $\bar{\rho}$, box size $L$ and grid size $N$.

astro-ph.CO↗

arXiv2023

Kinetic relaxation and Bose-star formation in multicomponent dark matter- I

Using wave kinetics, we estimate the emergence time-scale of gravitating Bose-Einstein condensates/Bose stars in the kinetic regime for a general multicomponent Schr\"{o}dinger-Poisson (SP) system. We identify some effects of the diffusion and friction pieces in the wave-kinetic Boltzmann equation (at leading order in perturbation theory) and provide estimates for the kinetic nucleation rate of condensates. We test our analysis using full $3+1$ dimensional simulations of multicomponent SP system. With an eye towards applications to multicomponent dark matter, we investigate two general cases in detail. First is a massive spin-$s$ field with $N=2s+1$ components (scalar $s=0$, vector $s=1$ and tensor $s=2$). We find that for a democratic population of different components, the condensation time-scale is $\tau_{(s)}\approx \tau_0\times N$, where $\tau_0$ is the condensation time scale for the scalar case. Second is the case of two scalars with different boson masses. In this case, we map-out how the condensation time depends on the ratios of their average mass densities and boson masses, revealing competition and assistance between components, and a guide towards which component condenses first. For instance, with $m_1 < m_2$ and not too disparate mass densities, we verify that the time scale of condensation of the first species quickly becomes independent of $m_2/m_1$, whereas for equal average number densities, the emergence time scale decreases with increasing $m_2/m_1$.

astro-ph.CO↗

arXiv2022

Sim2real for Reinforcement Learning Driven Next Generation Networks

The next generation of networks will actively embrace artificial intelligence (AI) and machine learning (ML) technologies for automation networks and optimal network operation strategies. The emerging network structure represented by Open RAN (O-RAN) conforms to this trend, and the radio intelligent controller (RIC) at the centre of its specification serves as an ML applications host. Various ML models, especially Reinforcement Learning (RL) models, are regarded as the key to solving RAN-related multi-objective optimization problems. However, it should be recognized that most of the current RL successes are confined to abstract and simplified simulation environments, which may not directly translate to high performance in complex real environments. One of the main reasons is the modelling gap between the simulation and the real environment, which could make the RL agent trained by simulation ill-equipped for the real environment. This issue is termed as the sim2real gap. This article brings to the fore the sim2real challenge within the context of O-RAN. Specifically, it emphasizes the characteristics, and benefits that the digital twins (DT) could have as a place for model development and verification. Several use cases are presented to exemplify and demonstrate failure modes of the simulations trained RL model in real environments. The effectiveness of DT in assisting the development of RL algorithms is discussed. Then the current state of the art learning-based methods commonly used to overcome the sim2real challenge are presented. Finally, the development and deployment concerns for the RL applications realisation in O-RAN are discussed from the view of the potential issues like data interaction, environment bottlenecks, and algorithm design.

cs.LG↗

arXiv2021

RLOps: Development Life-cycle of Reinforcement Learning Aided Open RAN

Radio access network (RAN) technologies continue to evolve, with Open RAN gaining the most recent momentum. In the O-RAN specifications, the RAN intelligent controllers (RICs) are software-defined orchestration and automation functions for the intelligent management of RAN. This article introduces principles for machine learning (ML), in particular, reinforcement learning (RL) applications in the O-RAN stack. Furthermore, we review the state-of-the-art research in wireless networks and cast it onto the RAN framework and the hierarchy of the O-RAN architecture. We provide a taxonomy for the challenges faced by ML/RL models throughout the development life-cycle: from the system specification to production deployment (data acquisition, model design, testing and management, etc.). To address the challenges, we integrate a set of existing MLOps principles with unique characteristics when RL agents are considered. This paper discusses a systematic model development, testing and validation life-cycle, termed: RLOps. We discuss fundamental parts of RLOps, which include: model specification, development, production environment serving, operations monitoring and safety/security. Based on these principles, we propose the best practices for RLOps to achieve an automated and reproducible model development process. At last, a holistic data analytics platform rooted in the O-RAN deployment is designed and implemented, aiming to embrace and fulfil the aforementioned principles and best practices of RLOps.

cs.NI↗