Reinforcement learning is a powerful approach to learn behaviour through interactions with
an environment. However, behaviours are usually learned in a purely reactive fashion …

Throughout this thesis, I develop the idea that the problem of learning good temporal
abstractions in reinforcement learning is intimately tied to a kind of representation learning …

Reinforcement Learning (RL) is a popular paradigm for sequential decision making under
uncertainty. A typical RL algorithm operates with only limited knowledge of the environment …

Reinforcement Learning (RL) has achieved remarkable success in various continuous
control tasks, such as robot manipulation and locomotion. Different to mainstream RL which …

The current reinforcement learning framework focuses exclusively on performance, often at
the expense of efficiency. In contrast, biological control achieves remarkable performance …

Recent successes in Reinforcement Learning have encouraged a fast-growing network of
RL researchers and a number of breakthroughs in RL research. As the RL community and …

We present temporally layered architecture (TLA), a biologically inspired system for
temporally adaptive distributed control. TLA layers a fast and a slow controller together to …

The idea of temporal abstraction, ie learning, planning and representing the world at
multiple time scales, has been a constant thread in AI research, spanning sub-fields from …

The temporal world is characterized by dynamic and variance. A lot of machine learning
algorithms are difficult to be applied to practical control applications directly, while …

This project is concerned with learning to take decisions in complex domains, in games in
particular. Previous work assumes that massive data resources are available for training, but …