AINeutralarXiv – CS AI · 10h ago5/10
🧠
NASDAQ: Normalized Observation Space Dynamics-Augmented Q-Learning
Researchers propose NASDAQ, a reinforcement learning framework that addresses performance degradation in low-dimensional observation tasks by normalizing observation spaces before dynamics prediction. The method balances reconstruction losses across observation dimensions and achieves competitive performance with faster training than existing model-based and self-predictive RL approaches.