Marek Cygan

sample efficiency (3) deep reinforcement learning (2) reinforcement learning (1) policy gradient (1) autoregressive generation (1) continuous control (1) model-based reinforcement learning (1) variance reduction (1) expected utility (1) optimistic exploration (1) temporal difference (1) mixture of expert (1) continuous action (1) parameter scaling (1) model-free algorithm (1) critic network (1) pessimistic policy (1) risk-aware reinforcement learning (1) pessimistic actor-critic (1) critic approximation error (1)