Reinforcement Learning with Perturbed Rewards
national conference on artificial intelligence, 2020.
Recent studies have shown that reinforcement learning (RL) models are vulnerable in various noisy scenarios. For instance, the observed reward channel is often subject to noise in practice (e.g., when rewards are collected through sensors), and is therefore not credible. In addition, for applications such as robotics, a deep reinforceme...More
PPT (Upload PPT)