Anders Jonsson

Average-reward option Q-learning achieves $\tilde{O}(1/\text{iterations})$ convergence, unlocking continual RL tasks previously intractable for option-based methods.

CG-CMARL scales to multi-agent scenarios without the exponential blowup in complexity, outperforming fixed reward-shaping methods in achieving optimal coordination.

Reinforcement learning can now handle active feature selection in high-dimensional datasets by intelligently pruning the feature search space and regularizing decision sequences, outperforming existing methods in accuracy and policy complexity.