8 papers
Calibrated Partial Resets: Preventing Policy Collapse in Continual Reinforcement Learning
Luc McCutcheon, Evangelos Chatzaroulas, Saber Fallah
Neural networks are hindered by accumulating dormant neurons and loss of expressivity throughout training, particularly in non-stationary data settings, such as continual supervise…
ANDRE: An Attention-based Neuro-symbolic Differentiable Rule Extractor for Inductive Logic Programming
Iman Sharifi, Peng Wei, Saber Fallah
Inductive Logic Programming (ILP) aims to learn interpretable first-order rules from data, but existing symbolic and neuro-symbolic approaches struggle to scale to noisy and probab…
Reinforcement Learning Enhancement Using Vector Semantic Representation and Symbolic Reasoning for Human-Centered Autonomous Emergency Braking
Vinal Asodia, Iman Sharifi, Saber Fallah
The problem with existing camera-based Deep Reinforcement Learning approaches is twofold: they rarely integrate high-level scene context into the feature representation, and they r…
Offline Reinforcement Learning using Human-Aligned Reward Labeling for Autonomous Emergency Braking in Occluded Pedestrian Crossing
Vinal Asodia, Barkin Dagda, Yinglong He +2
Effective leveraging of real-world driving datasets is crucial for enhancing the training of autonomous driving systems. While Offline Reinforcement Learning enables training auton…
Symbolic Imitation Learning: From Black-Box to Explainable Driving Policies
Iman Sharifi, Mustafa Yildirim, Saber Fallah
Current imitation learning approaches, predominantly based on deep neural networks (DNNs), offer efficient mechanisms for learning driving policies from real-world datasets. Howeve…
Towards Safe Autonomous Driving Policies using a Neuro-Symbolic Deep Reinforcement Learning Approach
Iman Sharifi, Mustafa Yildirim, Saber Fallah
The dynamic nature of driving environments and the presence of diverse road users pose significant challenges for decision-making in autonomous driving. Deep reinforcement learning…