Two new reinforcement learning papers accepted at ICLR 2024 Conference.
- [Spotlight✨] Robust Adversarial Reinforcement Learning via Bounded Rationality Curricula by Aryaman Reddi, Maximilian Tölle, Jan Peters, Georgia Chalvatzaki, Carlo D’Eramo
A novel approach for adversarial reinforcement learning that adopts a game-theoretical perspective based on bounded rationality to improve the robustness of obtained policies.
~ QARL
- Domain Randomization via Entropy Maximization by Gabriele Tiboni, Pascal Klink, Jan Peters, Tatiana Tommasi, Carlo D’Eramo, Georgia Chalvatzaki
A novel approach for domain randomization that directly maximizes the entropy of the dynamics randomizing distribution during training, while retaining generalization capabilities.
~ DORAEMON
Make sure to say hi 👋 to our wonderful students in Vienna.