AINeutralarXiv – CS AI · 6h ago6/10
🧠
Temporal Self-Imitation Learning
Researchers introduce Temporal Self-Imitation Learning (TSIL), a reinforcement learning framework that improves robot manipulation training by identifying and reusing efficient successful trajectories as self-supervision signals. The approach outperforms traditional reward-shaping methods across 15 long-horizon tasks by leveraging temporal efficiency as an intrinsic learning signal rather than relying solely on manually engineered rewards.