6 papers
Breaking the Epistemic Trap: Active Perception Under Compound Uncertainty
Chayan Banerjee, Ethan Goan
Deploying reinforcement learning in safety critical domains, from autonomous vehicles to medical decision support, is constrained by failures arising when systems encounter unfamil…
CWM: Contrastive World Models for Action Feasibility Learning in Embodied Agent Pipelines
Chayan Banerjee
A reliable action feasibility scorer is a critical bottleneck in embodied agent pipelines: before any planning or reasoning occurs, the agent must identify which candidate actions…
Sustainable Multi-Agent Crowdsourcing via Physics-Informed Bandits
Chayan Banerjee
Crowdsourcing platforms face a four-way tension between allocation quality, workforce sustainability, operational feasibility, and strategic contractor behaviour--a dilemma we form…
Physics-Informed Neuro-Symbolic Recommender System: A Dual-Physics Approach for Personalized Nutrition
Chayan Banerjee
Traditional e-commerce recommender systems primarily optimize for user engagement and purchase likelihood, often neglecting the rigid physiological constraints required for human h…
Reinforcement Learning Based Traffic Signal Design to Minimize Queue Lengths
Anirud Nandakumar, Chayan Banerjee, Lelitha Devi Vanajakshi
Efficient traffic signal control (TSC) is crucial for reducing congestion, travel delays, pollution, and for ensuring road safety. Traditional approaches, such as fixed signal cont…
Zero-Shot LLMs in Human-in-the-Loop RL: Replacing Human Feedback for Reward Shaping
Mohammad Saif Nazir, Chayan Banerjee
Reinforcement learning (RL) often struggles with reward misalignment, where agents optimize given rewards but fail to exhibit the desired behaviors. This arises when the reward fun…