arXiv subjects

Riya Singh

Publications and source records attributed to Riya Singh.

5 recordsLinked to original sources

arXiv2026

Homing through Reinforcement Learning

Homing and navigation are fundamental behaviors in biological systems that enable agents to reliably reach a target under uncertainty. We present a Reinforcement Learning (RL) framework to model adaptive homing in continuous two-dimensional domain. In this framework, the agent's state is given by its angular deviation from home, actions correspond to alignment or stochastic reorientation, and learning is driven by a radial-distance-based cost that penalizes motion away from the target, where the cost also acts as an effective signal guiding the agent towards the home. For a single self-propelled agent moving with constant speed, we find that the mean homing time $\langle T_{\mathrm{home}} \rangle$ exhibits a non-monotonic dependence on the rotational diffusion strength $D_r$, with an optimal noise level $D_r^\ast$, revealing a subtle interplay between exploration and goal-directed correction. Extending to two agents with soft repulsion, one agent consistently reaches home faster than the other, while in multi-agents system, repulsion ensures separation and the fastest agent becomes progressively faster as group size increases. Finally, we have compared the homing time obtained from the RL agent with that of an Active Brownian Particle (ABP) with resetting and a pure ABP (without resetting) under identical conditions. The RL-based agent consistently achieves shorter homing times with trajectories that are less noisy and more directed than both cases, while the pure ABP typically continues wandering near the target without reliable localization. Our results show that cost-driven learning, stochastic reorientation, and inter-agent interactions enable efficient adaptive navigation, linking individual and collective homing. This RL framework captures key biological features such as feedback-based route learning, randomness to escape unfavorable orientations, and mutual coordination.

cond-mat.soft↗

arXiv2026

Two-step Authentication: Multi-biometric System Using Voice and Facial Recognition

We present a cost-effective two-step authentication system that integrates face identification and speaker verification using only a camera and microphone available on common devices. The pipeline first performs face recognition to identify a candidate user from a small enrolled group, then performs voice recognition only against the matched identity to reduce computation and improve robustness. For face recognition, a pruned VGG-16 based classifier is trained on an augmented dataset of 924 images from five subjects, with faces localized by MTCNN; it achieves 95.1% accuracy. For voice recognition, a CNN speaker-verification model trained on LibriSpeech (train-other-360) attains 98.9% accuracy and 3.456% EER on test-clean. Source code and trained models are available at https://github.com/NCUE-EE-AIAL/Two-step-Authentication-Multi-biometric-System.

cs.CV↗

arXiv2025

DeepRetro: Retrosynthetic Pathway Discovery using Iterative LLM Reasoning

The synthesis of complex natural products remains one of the grand challenges of organic chemistry. We present DeepRetro, a major advancement in computational retrosynthesis that enables the discovery of viable synthetic routes for complex molecules typically considered beyond the reach of existing retrosynthetic methods. DeepRetro is a novel, open-source framework that tightly integrates large language models (LLMs), traditional retrosynthetic engines, and expert human feedback in an iterative design loop. Prior approaches rely solely on template-based methods or unconstrained LLM outputs. In contrast, DeepRetro combines the precision of template-based methods with the generative flexibility of LLMs, controlled by rigorous chemical validity checks and enhanced by recursive refinement. This hybrid system dynamically explores and revises synthetic pathways, guided by both algorithmic checks and expert chemist feedback through an interactive user interface. While DeepRetro achieves strong performance on standard retrosynthesis benchmarks, its true strength lies in its ability to propose novel, viable pathways to highly complex natural products-targets that have historically eluded automated planning. Through detailed case studies, we illustrate how this approach enables new routes for total synthesis and facilitates human-machine collaboration in organic chemistry. Beyond retrosynthesis, DeepRetro represents a working model for how to leverage LLMs in scientific discovery. We provide a transparent account of the system's design, algorithms, and human-feedback loop, enabling broad adaptation across scientific domains. By releasing DeepRetro as an open-source tool, we aim to empower chemists to tackle increasingly ambitious synthetic targets, accelerating progress in drug discovery, materials design, and beyond.

q-bio.QM↗

arXiv2024

Adaptive dynamics of Ising spins in one dimension leveraging Reinforcement Learning

A one-dimensional flocking model using active Ising spins is studied, where the system evolves through the reinforcement learning approach \textit{via} defining state, action, and cost function for each spin. The orientation of spin with respect to its neighbouring spins defines its state. The state of spin is updated by altering its spin orientation in accordance with the $\varepsilon$-greedy algorithm (action) and selecting a finite step from a uniform distribution to update position. The $\varepsilon$ parameter is analogous to the thermal noise in the system. The cost function addresses cohesion among the spins. By exploring the system in the plane of the self-propulsion speed and $\varepsilon$ parameter, four distinct phases are found: disorder, flocking, flipping, and oscillatory. In the flipping phase, a condensed flock reverses its direction of motion stochastically. The mean reversal time $\langle T \rangle $ exponentially decays with $\varepsilon$. A new phase, an oscillatory phase, is also found, which is a chaotic phase with a positive Lyapunov exponent. The findings obtained from the reinforcement learning approach for the active Ising model system exhibit similarities with the outcomes of other conventional techniques, even without defining any explicit interaction among the spins.

cond-mat.soft↗

arXiv2021

Correlated Single- and Few-Electron Backgrounds Milliseconds after Interactions in Dual-Phase Liquid Xenon Time Projection Chambers

We characterize single- and few-electron backgrounds that are observed in dual-phase liquid xenon time projection chambers at timescales greatly exceeding a maximum drift time after an interaction. These instrumental backgrounds limit a detector's sensitivity to dark matter and cosmogenic neutrinos. Using the ~150g liquid xenon detector at Purdue University, we investigate how these backgrounds, produced after 122keV $^{57}$Co Compton interactions, behave under different detector conditions. We find that the rates of single- and few-electron signals follow power-laws with time after the interaction. We observe linearly increasing rates with increased extraction field, and increased rates in the single-electron background with increased drift field. Normalizing the rates to the primary interaction's measured ionization signal, the rates increase linearly with the depth of the interaction. We test the hypothesis that infrared photons (1550nm) would stimulate and accelerate electron emission via photodetachment from impurities, but find that even 1 Watt of infrared light fails to reduce these backgrounds. We thus provide a characterization that can inform background models for low-energy rare event searches.

physics.ins-det↗