#policy optimization
2 articles
Latest
Research
Seven Papers Advance RL Policy Optimization and Inverse Learning
New methods address bootstrapping error, inverse reward inference, and offline learning challenges with distributional and theoretical approaches.
about 10 hours ago9 min read
Research
Eight Papers Advance RL Training: Imagination, Reward Models, Optimizer Theory
New methods address dynamics simulation, calibration bias, gradient stability, and weak feedback signals in reinforcement learning systems.
1 day ago10 min read