arXiv subjects

Samuel Tovey

Publications and source records attributed to Samuel Tovey.

At least 19 recordsLinked to original sources

arXiv2026

Multi-modal transformer for signal classification in nanopore blockade experiments

Nanopore devices have emerged as powerful tools for single-molecule sensing, with potential for rapid, portable diagnostics. They detect changes in ionic current as analytes enter nanometer-scale pores, providing a means of identifying diverse biomarkers from their characteristic signal patterns. However, these signals are highly complex, and reliably assigning them to specific molecules remains a major challenge. Here, we address this by introducing a multi-modal deep learning architecture that jointly processes multiple signal representations, including raw time-series data, wavelet-based images, and static feature vectors. Our approach surpasses existing methods by more than 10 percentage points on a 42-peptide benchmark and transfers to a 20-amino-acid dataset with near-perfect accuracy. The model integrates complementary information from these representations, with attention analysis showing that the time-series and wavelet-image inputs emphasize different features of the same event. Together, these results demonstrate the potential of machine learning to enable robust, high-accuracy molecular identification with nanopore sensors.

cs.LG↗

arXiv2026

From Rubble Simulation to Active Magnetic Mapping: Quantum Sensing for Disaster Response

Locating survivors of building collapses within the first 72 hours is a critical challenge in disaster response, and existing sensing modalities provide only partial information about the structure beneath the rubble. This paper proposes drone-based quantum magnetometry as a complementary modality and develops a simulation pipeline spanning rubble physics, sensor-array deployment, and active spatial reconstruction. We use Unreal Engine to generate a steel-reinforced concrete parking-garage collapse and compute the induced magnetic field via a per-triangle dipole approximation, establishing that meaningful magnetic structure is recoverable in the sub-pT to sub-nT range from roughly 1 m above the roofline. Then, we feed sparse multi-sensor samples into a Gaussian Process Regression back-end driven by Bayesian active sampling and validate the pipeline across multiple independent collapse realizations; a three-sensor array optimizes the trade-off between gradient resolution and UAV payload constraints, and active sampling reaches peak structural correlation in roughly $100$ samples. Together, these results indicate that quantum-grade sensing could become a useful tool for drone-based structural analysis and potentially void detection in collapsed buildings.

cs.RO↗

arXiv2026

Reinforcement Learning Enables Autonomous Microrobot Navigation and Intervention in Simulated Blood Capillaries

Autonomous microrobots navigating biological vasculature could enable targeted drug delivery and thrombolysis, yet training control policies for realistic environments remains an open challenge. Prior reinforcement learning (RL) studies of microrobotic navigation have been limited to idealized geometries that omit complex hydrodynamic flow fields, confined branching structures, and dense cellular obstacles found in vivo. Here, we develop a physically grounded simulation of a blood capillary network, incorporating realistic hydrodynamic flow fields, explicit red blood cell dynamics, and anatomically derived branching geometry, and train deep RL agents to navigate it via chemotaxis. We systematically map the physical limits of navigation across robot size and swimming speed, revealing a forbidden regime where Brownian motion and flow overcome propulsion. Successful agents independently discover multiple universal strategy types, including run-and-rotate and energy-efficient search-and-sit policies, regardless of robot parameters. Without retraining, these agents perform targeted blocking and unblocking of capillary flow, restoring throughput to healthy baseline levels. These results establish RL as a viable framework for developing autonomous microrobotic intervention strategies in complex biological environments.

cs.RO↗

arXiv2026

Spectral Reach: Understanding Neural Scaling as Progress into the Spectral Tail

Neural scaling laws describe predictable power-law relationships between model size, dataset size, compute, and performance. While these laws guide the development of modern foundation models, the mechanisms underpinning them remain poorly understood, in part due to the absence of scalable analysis tools. To close this gap, we introduce "spectral position": a scalable measure of which eigenvalues of the empirical neural tangent kernel (eNTK) currently drive loss reduction. Applying this measure to scaling experiments, we find that spectral position decreases throughout training: learning shifts from dominant eigenmodes into the spectral tail. Larger models reach further into the tail than smaller models, revealing a size-dependent capacity we call "spectral reach". This suggests why larger models achieve lower losses: they sustain learning on weak spectral signals inaccessible to smaller models. We further identify feature learning as a key enabler of spectral reach. It adaptively amplifies gradient magnitudes as learning advances, sustaining progress where frozen representations stall. This points to concrete interventions through architecture and optimizer design.

cs.LG↗

arXiv2026

Sequential Spatiotemporal Magnetic-Field Reconstruction via Quantum Hamiltonian Learning with NV-Center Spin-1 Hamiltonians

We propose a quantum-Hamiltonian-learning-based sequential reconstruction framework for dynamic two-dimensional magnetic-field maps using a local likelihood model derived from a nitrogen-vacancy-center spin-1 Hamiltonian. Local measurements are generated through nitrogen-vacancy spin dynamics governed by local magnetic-field values and a shared dipolar coupling parameter, rather than by direct observation of the latent field. Sequential Bayesian updates over overlapping scan windows are combined with temporal posterior propagation to reconstruct the evolving field. Numerical proof-of-concept experiments on controlled synthetic maze-like magnetic-field sequences show that the proposed method reconstructs the dominant spatial structure of the tested field class, achieving a final-frame RMSE of $7.037\times10^{-7}\,\mathrm{T}$. Adaptive diagnostics show decreasing expected information gain and stable local convergence, while Fisher-information and leakage diagnostics reveal a sensitivity--leakage tradeoff under long-interrogation controls. Combined horizontal and vertical scans yield better reconstruction than single-direction acquisition in the tested setting. In contrast, the shared coupling parameter $J$ is only partially identifiable: its posterior becomes narrow but remains frame-dependent and biased. At the final checkpoint, $J_{\rm std}=87.0\,\mathrm{Hz}$, close to a finite-time product-state reference benchmark of $73.3\,\mathrm{Hz}$, while remaining $3.35\times$ above a gain-extrapolated ideal-state benchmark. The posterior mean remains biased by $326.9\,\mathrm{Hz}$, indicating that posterior concentration alone does not imply unbiased coupling recovery. These results demonstrate feasibility for the tested structured field class and identify coupling estimation as the main identifiability bottleneck.

quant-ph↗

arXiv2026

Reservoir computing from collective dynamics of active colloidal oscillators

Physical reservoir computing is a computational framework that offers an energy- and computation-efficient alternative to conventional training of neural networks. In reservoir computing, input signals are mapped into the high-dimensional dynamics of a nonlinear system, and only a simple readout layer is trained. In most physical implementations, the interactions that give rise to the dynamics cannot be tuned directly and high dimensionality is typically achieved through time-multiplexing, which can limit flexibility and efficiency. Here we introduce a reservoir composed of hundreds of hydrodynamically coupled active colloidal oscillators forming a fully parallel physical reservoir and whose coupling strength and fading-memory time can be tuned in situ. The collective dynamics of the active oscillators allow accurate predictions of chaotic time series from single reservoir readouts without time-multiplexing. We further demonstrate real-time detection of subtle hidden anomalies that preserve all instantaneous statistical properties of the signal yet disrupt its underlying temporal correlations. These results establish interacting active colloids as a reconfigurable platform for physical computation and edge-integrated intelligent sensing for model-free detection of irregularities in complex time signals.

cond-mat.soft↗

arXiv2025

Deep Learning-Driven Peptide Classification in Biological Nanopores

Nanopore-based single-molecule sensing is a promising route to fast, low-cost disease diagnosis and protein sequencing: as an analyte such as a peptide or protein traverses a nanoscale pore, it modulates the ionic current, producing a resistive pulse whose signature is determined by the analyte's structure and its interactions with the pore. Translating these signatures into reliable molecular identities, however, is an open problem well suited for machine learning, as the signals are noisy, suffer from variations due to experimental conditions, and are difficult to featurize, which has so far limited classification accuracy. Here we translate the peptide identification problem into an image-classification task by transforming each resistive pulse into a scaleogram via the continuous wavelet transform, a representation that jointly encodes amplitude, frequency, and time in a form well suited for deep convolutional models. On a dataset of 42 peptides, recorded as six separate peptide ladders, this approach reaches a macro-averaged classification accuracy of $82\,\%$ on held-out events, an improvement of $8.6$ percentage points over the descriptor-based approach previously reported for the same dataset. We further show that the trained models tolerate substantial compression, retaining their accuracy with half of their weights set to zero and under 8-bit quantization, a prerequisite for deploying trained classifiers on embedded sensing hardware. Our results demonstrate how physically motivated signal representations can make complex single-molecule data tractable for modern learning algorithms, a step on the path towards point-of-care peptide and protein diagnostics.

cs.LG↗

arXiv2025

Beyond Scaling Curves: Internal Dynamics of Neural Networks Through the NTK Lens

Scaling laws offer valuable insights into the relationship between neural network performance and computational cost, yet their underlying mechanisms remain poorly understood. In this work, we empirically analyze how neural networks behave under data and model scaling through the lens of the neural tangent kernel (NTK). This analysis establishes a link between performance scaling and the internal dynamics of neural networks. Our findings of standard vision tasks show that similar performance scaling exponents can occur even though the internal model dynamics show opposite behavior. This demonstrates that performance scaling alone is insufficient for understanding the underlying mechanisms of neural networks. We also address a previously unresolved issue in neural scaling: how convergence to the infinite-width limit affects scaling behavior in finite-width models. To this end, we investigate how feature learning is lost as the model width increases and quantify the transition between kernel-driven and feature-driven scaling regimes. We identify the maximum model width that supports feature learning, which, in our setups, we find to be more than ten times smaller than typical large language model widths.

cs.LG↗

arXiv2025

Quantum vs. classical: A comprehensive benchmark study for predicting time series with variational quantum machine learning

Variational quantum machine learning algorithms have been proposed as promising tools for time series prediction, with the potential to handle complex sequential data more effectively than classical approaches. However, their practical advantage over established classical methods remains uncertain. In this work, we present a comprehensive benchmark study comparing a range of variational quantum algorithms and classical machine learning models for time series forecasting. We evaluate their predictive performance on three chaotic systems across 27 time series prediction tasks of varying complexity, and ensure a fair comparison through extensive hyperparameter optimization. Our results indicate that, in many cases, quantum models struggle to match the accuracy of simple classical counterparts of comparable complexity. Furthermore, we analyze the predictive performance relative to the model complexity and discuss the practical limitations of variational quantum algorithms for time series forecasting.

quant-ph↗

arXiv2024

Collective variables of neural networks: empirical time evolution and scaling laws

This work presents a novel means for understanding learning dynamics and scaling relations in neural networks. We show that certain measures on the spectrum of the empirical neural tangent kernel, specifically entropy and trace, yield insight into the representations learned by a neural network and how these can be improved through architecture scaling. These results are demonstrated first on test cases before being shown on more complex networks, including transformers, auto-encoders, graph neural networks, and reinforcement learning studies. In testing on a wide range of architectures, we highlight the universal nature of training dynamics and further discuss how it can be used to understand the mechanisms behind learning in neural networks. We identify two such dominant mechanisms present throughout machine learning training. The first, information compression, is seen through a reduction in the entropy of the NTK spectrum during training, and occurs predominantly in small neural networks. The second, coined structure formation, is seen through an increasing entropy and thus, the creation of structure in the neural network representations beyond the prior established by the network at initialization. Due to the ubiquity of the latter in deep neural network architectures and its flexibility in the creation of feature-rich representations, we argue that this form of evolution of the network's entropy be considered the onset of a deep learning regime.

cs.LG↗

arXiv2024

Peptide Classification from Statistical Analysis of Nanopore Translocation Experiments

Protein characterization using nanopore-based devices promises to be a breakthrough method in basic research, diagnostics, and analytics. Current research includes the use of machine learning to achieve this task. In this work, a comprehensive statistical analysis of nanopore current signals is performed and demonstrated to be sufficient for classifying up to 42 peptides with over 70 % accuracy. Two sets of features, the statistical moments and the catch22 set, are compared both in their representations and after training small classifier neural networks. We demonstrate that complex features of the events, captured in both the catch22 set and the central moments, are key in classifying peptides with otherwise similar mean currents. These results highlight the efficacy of purely statistical analysis of nanopore data and suggest a path forward for more sophisticated classification techniques.

physics.bio-ph↗

arXiv2024

SwarmRL: Building the Future of Smart Active Systems

This work introduces SwarmRL, a Python package designed to study intelligent active particles. SwarmRL provides an easy-to-use interface for developing models to control microscopic colloids using classical control and deep reinforcement learning approaches. These models may be deployed in simulations or real-world environments under a common framework. We explain the structure of the software and its key features and demonstrate how it can be used to accelerate research. With SwarmRL, we aim to streamline research into micro-robotic control while bridging the gap between experimental and simulation-driven sciences. SwarmRL is available open-source on GitHub at https://github.com/SwarmRL/SwarmRL.

cs.RO↗

arXiv2024

Generating Quantum Reservoir State Representations with Random Matrices

We demonstrate a novel approach to reservoir computation measurements using random matrices. We do so to motivate how atomic-scale devices could be used for real-world computational applications. Our approach uses random matrices to construct reservoir measurements, introducing a simple, scalable means of generating state representations. In our studies, two reservoirs, a five-atom Heisenberg spin chain and a five-qubit quantum circuit, perform time series prediction and data interpolation. The performance of the measurement technique and current limitations are discussed in detail, along with an exploration of the diversity of measurements provided by the random matrices. In addition, we explore the role of reservoir parameters such as coupling strength and measurement dimension, providing insight into how these learning machines could be automatically tuned for different problems. This research highlights the use of random matrices to measure simple quantum reservoirs for natural learning devices, and outlines a path forward for improving their performance and experimental realization.

quant-ph↗

arXiv2024

Emergence of Chemotactic Strategies with Multi-Agent Reinforcement Learning

Reinforcement learning (RL) is a flexible and efficient method for programming micro-robots in complex environments. Here we investigate whether reinforcement learning can provide insights into biological systems when trained to perform chemotaxis. Namely, whether we can learn about how intelligent agents process given information in order to swim towards a target. We run simulations covering a range of agent shapes, sizes, and swim speeds to determine if the physical constraints on biological swimmers, namely Brownian motion, lead to regions where reinforcement learners' training fails. We find that the RL agents can perform chemotaxis as soon as it is physically possible and, in some cases, even before the active swimming overpowers the stochastic environment. We study the efficiency of the emergent policy and identify convergence in agent size and swim speeds. Finally, we study the strategy adopted by the reinforcement learning algorithm to explain how the agents perform their tasks. To this end, we identify three emerging dominant strategies and several rare approaches taken. These strategies, whilst producing almost identical trajectories in simulation, are distinct and give insight into the possible mechanisms behind which biological agents explore their environment and respond to changing conditions.

physics.bio-ph↗

arXiv2024

Emergence of Accurate Atomic Energies from Machine Learned Noble Gas Potentials

The quantum theory of atoms in molecules (QTAIM) gives access to well-defined local atomic energies. Due to their locality, these energies are potentially interesting in fitting atomistic machine learning models as they inform about physically relevant properties. However, computationally, quantum-mechanically accurate local energies are notoriously difficult to obtain for large systems. Here, we show that by employing semi-empirical correlations between different components of the total energy, we can obtain well-defined local energies at a moderate cost. We employ this methodology to investigate energetics in noble liquids or argon, krypton, and their mixture. Instead of using these local energies to fit atomistic models, we show how well these local energies are reproduced by machine-learned models trained on the total energies. The results of our investigation suggest that smaller neural networks, trained only on the total energy of an atomistic system, are more likely to reproduce the underlying local energy partitioning faithfully than larger networks. Furthermore, we demonstrate that networks more capable of this energy decomposition are, in turn, capable of transferring to previously unseen systems. Our results are a step towards understanding how much physics can be learned by neural networks and where this can be applied, particularly how a better understanding of physics aids in the transferability of these neural networks.

cond-mat.soft↗

arXiv2024

ZnTrack -- Data as Code

The past decade has seen tremendous breakthroughs in computation and there is no indication that this will slow any time soon. Machine learning, large-scale computing resources, and increased industry focus have resulted in rising investments in computer-driven solutions for data management, simulations, and model generation. However, with this growth in computation has come an even larger expansion of data and with it, complexity in data storage, sharing, and tracking. In this work, we introduce ZnTrack, a Python-driven data versioning tool. ZnTrack builds upon established version control systems to provide a user-friendly and easy-to-use interface for tracking parameters in experiments, designing workflows, and storing and sharing data. From this ability to reduce large datasets to a simple Python script emerges the concept of Data as Code, a core component of the work presented here and an undoubtedly important concept as the age of computation continues to evolve. ZnTrack offers an open-source, FAIR data compatible Python package to enable users to harness these concepts of the future.

cs.SE↗

arXiv2023

Generating Minimal Training Sets for Machine Learned Potentials

This letter presents a novel approach for identifying uncorrelated atomic configurations from extensive data sets with a non-standard neural network workflow known as random network distillation (RND) for training machine-learned inter-atomic potentials (MLPs). This method is coupled with a DFT workflow wherein initial data is generated with cheaper classical methods before only the minimal subset is passed to a more computationally expensive ab initio calculation. This benefits training not only by reducing the number of expensive DFT calculations required but also by providing a pathway to the use of more accurate quantum mechanical calculations for training. The method's efficacy is demonstrated by constructing machine-learned inter-atomic potentials for the molten salts KCl and NaCl. Our RND method allows accurate models to be fit on minimal data sets, as small as 32 configurations, reducing the required structures by at least one order of magnitude compared to alternative methods.

physics.comp-ph↗

arXiv2023

Environmental effects on emergent strategy in micro-scale multi-agent reinforcement learning

Multi-Agent Reinforcement Learning (MARL) is a promising candidate for realizing efficient control of microscopic particles, of which micro-robots are a subset. However, the microscopic particles' environment presents unique challenges, such as Brownian motion at sufficiently small length-scales. In this work, we explore the role of temperature in the emergence and efficacy of strategies in MARL systems using particle-based Langevin molecular dynamics simulations as a realistic representation of micro-scale environments. To this end, we perform experiments on two different multi-agent tasks in microscopic environments at different temperatures, detecting the source of a concentration gradient and rotation of a rod. We find that at higher temperatures, the RL agents identify new strategies for achieving these tasks, highlighting the importance of understanding this regime and providing insight into optimal training strategies for bridging the generalization gap between simulation and reality. We also introduce a novel Python package for studying microscopic agents using reinforcement learning (RL) to accompany our results.

physics.bio-ph↗