In the paper, we propose a class of efficient momentum-based policy gradient methods for
the model-free reinforcement learning, which use adaptive learning rates and do not require …

In reinforcement learning, importance sampling is a widely used method for evaluating an
expectation under the distribution of data of one policy when the data has in fact been …

Applying probabilistic models to reinforcement learning (RL) enables the uses of powerful
optimisation tools such as variational inference in RL. However, existing inference …

In this thesis, we develop a comprehensive account of the expressive power, modelling
efficiency, and performance advantages of so-called trading agents (ie, Deep Soft Recurrent …

We introduce a novel perspective on Bayesian reinforcement learning (RL); whereas
existing approaches infer a posterior over the transition distribution or Q-function, we …

Reinforcement learning is able to obtain generalized low-level robot policies on diverse
robotics datasets in embodied learning scenarios, and Transformer has been widely used to …

We propose a general framework for policy representation for reinforcement learning tasks.
This framework involves finding a low-dimensional embedding of the policy on a …

Many reinforcement learning approaches rely on temporal-difference (TD) learning to learn
a critic. However, TD-learning updates can be high variance. Here, we introduce a model …

While often stated as an instance of the likelihood ratio trick [Rubinstein, 1989], the original
policy gradient theorem [Sutton, 1999] involves an integral over the action space. When this …

Policy Optimization (PO) is a family of reinforcement learning algorithms that is particularly
suited to real-world control tasks due to its ability of managing high-dimensional decision …