5 papers
The Informational Cost of Agency: A Bounded Measure of Interaction Efficiency for Deployed Reinforcement Learning
Wael Hafez, Cameron Reid, Amit Nazeri
Deployed reinforcement learning systems lack a principled runtime reliability theory. We close this gap by introducing Bipredictability, P, a closed form information theoretic metr…
Token Statistics Reveal Conversational Drift in Multi-turn LLM Interaction
Wael Hafez, Amir Nazeri
Large language models, LLMs, are increasingly deployed in multiturn settings where earlier responses shape later ones, making reliability dependent on whether a conversation remain…
Information-Theoretic Framework for Self-Adapting Model Predictive Controllers
Wael Hafez, Amir Nazeri
Model Predictive Control (MPC) is a vital technique for autonomous systems, like Unmanned Aerial Vehicles (UAVs), enabling optimized motion planning. However, traditional MPC strug…
Mutual Information Tracks Policy Coherence in Reinforcement Learning
Cameron Reid, Wael Hafez, Amirhossein Nazeri
Reinforcement Learning (RL) agents deployed in real-world environments face degradation from sensor faults, actuator wear, and environmental shifts, yet lack intrinsic mechanisms t…
Entropy-Based Non-Invasive Reliability Monitoring of Convolutional Neural Networks
Amirhossein Nazeri, Wael Hafez
Convolutional Neural Networks (CNNs) have become the foundation of modern computer vision, achieving unprecedented accuracy across diverse image recognition tasks. While these netw…