UnlikeGaussian policies, the log-likelihood indiffusion policies isinaccessible; thus this entropy term is nontrivial. Moreover, to reduce the large variance of diffusion policies, we also develop an efficient behavior policy through action selection. This can further improve its sample efficiency during online interaction.

Wemodel 34], whichisformallyG = hS, U, Pi. S is thestatespacet, every i 2 A {1,..., n} choosesui 2 U which action u 2 U Un. P(s0|s,u): S U S!

Neural Information Processing Systems http://nips.cc/

Recent advances in constrained reinforcement learning (RL) have endowed reinforcement learning with certain safety guarantees.

We propose information gating as a way to learn parsimonious representations that identify the minimal information required for a task.

Moreover, our algorithm is complementary to existing continual learning strategies, such as MER and GEM.

Distributional Reinforcement Learning (RL) differs from traditional RL in that, rather than the expectation of total returns, it estimates distributions and has achieved state-of-the-art performance on Atari Games. The key challenge in practical distributional RL algorithms lies in how to parameterize estimated distributions so as to better approximate the true continuous distribution.

Neural Information Processing Systems http://nips.cc/