Subramanya Nageshrao

Generating Socially Acceptable Perturbations for Efficient Evaluation of Autonomous Vehicles

2020

Deep reinforcement learning methods have been considered and implemented for autonomous vehicle's decisionmaking in recent years. A key issue is that deep neural networks can be fragile to adversarial attacks through unseen inputs, and thus …

Assured learning‐enabled autonomy: A metacognitive reinforcement learning framework

2021 · International Journal of Adaptive Control and Signal Processing

Summary Reinforcement learning (RL) agents with pre‐specified reward functions cannot provide guaranteed safety across variety of circumstances that an uncertain system might encounter. To guarantee performance while assuring satisfaction of safety constraints across variety of …