Preprint
Oct 2026
Interpretable Meta-Policies for Neural Decision Making
Deep Reinforcement Learning remains difficult to interpret and align due to the black-box nature of neural networks. While (neuro-)symbolic approaches improve transparency by incorporating symbolic reasoning, applying them directly to low-level actions results in overly complex, unreadable policies. We introduce NEXUS,...
Raban Emunds, Jannis Blüml, Quentin Delfosse et al.
· 0 citations