7 papers
JEDI: Joint Embedding Diffusion World Model for Online Model-Based Reinforcement Learning
Jing Yu Lim, Rushi Shah, Zarif Ikram +4
Diffusion world models have recently become competitive for online model-based reinforcement learning, but current approaches expose a tension: pixel diffusion is effective but com…
Performance Asymmetry in Model-Based Reinforcement Learning
Jing Yu Lim, Rushi Shah, Zarif Ikram +4
Recently, Model-Based Reinforcement Learning (MBRL) have achieved super-human level performance on the Atari100k benchmark on average. However, we discover that conventional aggreg…
Hierarchical Molecular Representation Learning via Fragment-Based Self-Supervised Embedding Prediction
Jiele Wu, Haozhe Ma, Zhihan Guo +2
Graph self-supervised learning (GSSL) has demonstrated strong potential for generating expressive graph embeddings without the need for human annotations, making it particularly va…
Centralized Reward Agent for Knowledge Sharing and Transfer in Multi-Task Reinforcement Learning
Haozhe Ma, Zhengding Luo, Thanh Vinh Vo +2
Reward shaping is effective in addressing the sparse-reward challenge in reinforcement learning (RL) by providing immediate feedback through auxiliary, informative rewards. Based o…
Exploration by Random Reward Perturbation
Haozhe Ma, Guoji Fu, Zhengding Luo +2
We introduce Random Reward Perturbation (RRP), a novel exploration strategy for reinforcement learning (RL). Our theoretical analyses demonstrate that adding zero-mean noise to env…
Causal Policy Learning in Reinforcement Learning: Backdoor-Adjusted Soft Actor-Critic
Thanh Vinh Vo, Young Lee, Haozhe Ma +2
Hidden confounders that influence both states and actions can bias policy learning in reinforcement learning (RL), leading to suboptimal or non-generalizable behavior. Most RL algo…