5 papers
Beyond Isolation: Unlocking Reinforcement Learning Component Synergy for Sample-Efficient Continuous Control
Qi Zhao, Guozheng Ma, Yilun Kong +9
Reinforcement learning systems are significantly more complex than other machine learning paradigms due to inherent properties, causing RL system design to jointly account for many…
Better, Faster: Harnessing Self-Improvement in Large Reasoning Models
Qihuang Zhong, Liang Ding, Juhua Liu +3
Self-improvement training enables the large reasoning models (LRMs) to improve themselves by self-generating reasoning trajectories as training data without external supervision. H…
ROAR: Retrieval Opportunity-Aware Refinement for Zero-Shot Time Series Forecasting
Jinjin Chi, Lei Feng, Lulu Zhang +7
Retrieval augmentation provides time series forecasters with historical continuations, yet even candidates that outperform the base forecast may fail to improve the final predictio…
What Makes Value Learning Efficient in Residual Reinforcement Learning?
Guozheng Ma, Lu Li, Haoyu Wang +3
Residual reinforcement learning (RL) enables stable online refinement of expressive pretrained policies by freezing the base and learning only bounded corrections. However, value l…
Rethinking the Role of Dynamic Sparse Training for Scalable Deep Reinforcement Learning
Guozheng Ma, Lu Li, Zilin Wang +4
Scaling neural networks has driven breakthrough advances in machine learning, yet this paradigm fails in deep reinforcement learning (DRL), where larger models often degrade perfor…