The following two papers from SISREL were accepted to ICML 2023.
Woojun Kim, Jeonghye Kim and Youngchul Sung, "LESSON: Learning to integrate exploration strategies for reinforcement learning via an option framework," to be presented at International Conference on Machine Learning (ICML), Hawaii, HI, Jul. 2023 (Woojun and Jeonghye are the co-first authors.)
Woojun Kim and Youngchul Sung, "An adaptive entropy regularizatio framework for multi-agent reinforcement learning," to be presented at International Conference on Machine Learning (ICML), Hawaii, HI, Jul. 2023
The first paper presents a new framework to unify multiple exploration strategies for best performance of reinforcement learning and the second paper presents a new way to achieve exploration-exploitation trade-off in multi-agent reinforcement learning based on varying individual agent's targent entropy.