Riccardo De Santi

non-markovian policy (2) convex reinforcement learning (2) finite trial (2) state distribution (2) markov decision process (1) model-based reinforcement learning (1) optimal policy (1) approximation error (1) regret bound (1) empirical distribution (1) deterministic policy (1) systematic generalization (1) causal model (1) planning task (1) state entropy (1) risk-averse reinforcement learning (1) causal inference (1) empirical state distribution (1) reinforcement learning (1) sample efficiency (1)