Safety
K-Score: Kalman Filter as a Principled Alternative to Reward Normalization in Reinforcement Learning
arXiv:2604.23056v1 Announce Type: cross Abstract: We propose a simple yet effective alternative to reward normalization in policy gradient reinforcement learning by integrating a 1D Kalman filter for
arXiv:2604.23056v1 Announce Type: cross Abstract: We propose a simple yet effective alternative to reward normalization in policy gradient reinforcement learning by integrating a 1D Kalman filter for online reward estimation. Instead of relying on fixed heuristics, our method recursively estimates the latent reward mean, smoothing high-variance returns and adapting to non-stationary environments. This approach incurs minimal overhead and requires no modification to existing policy architectures. Experiments on extit{LunarLander} and extit{CartPole} demonstrate that Kalman-filtered rewards significantly accelerate convergence and reduce training variance compared to standard normalization techniques. Code is available at https://github.com/Sumxiaa/Kalman_Normalization.
Related
- EVPO: Explained Variance Policy Optimization for Adaptive Critic Utilization in LLM Post-Training
- SCOPE: Signal-Calibrated On-Policy Distillation Enhancement with Dual-Path Adaptive Weighting
- Discrete Flow Matching Policy Optimization
- DVPO: Distributional Value Modeling-based Policy Optimization for LLM Post-Training
Source: arXiv cs.AI | 2026-04-28