2 papers
cs.LG2025
Off-Policy Actor-Critic for Adversarial Observation Robustness: Virtual Alternative Training via Symmetric Policy Evaluation
Kosuke Nakanishi, Akihiro Kubo, Yuji Yasui +1
Recently, robust reinforcement learning (RL) methods designed to handle adversarial input observations have received significant attention, motivated by RL's inherent vulnerabiliti…
cs.LG2024
Robust off-policy Reinforcement Learning via Soft Constrained Adversary
Kosuke Nakanishi, Akihiro Kubo, Yuji Yasui +1
Recently, robust reinforcement learning (RL) methods against input observation have garnered significant attention and undergone rapid evolution due to RL's potential vulnerability…