In this article, we aim to provide a literature review of different formulations and approaches
to continual reinforcement learning (RL), also known as lifelong or non-stationary RL. We …

Uncertainty is ubiquitous in games, both in the agents playing games and often in the games
themselves. Working with uncertainty is therefore an important component of successful …

Developing a general algorithm that learns to solve tasks across a wide range of
applications has been a fundamental challenge in artificial intelligence. Although current …

Deep reinforcement learning (RL) algorithms are predominantly evaluated by comparing
their relative performance on a large suite of tasks. Most published results on deep RL …

The study of zero-shot generalisation (ZSG) in deep Reinforcement Learning (RL) aims to
produce RL algorithms whose policies generalise well to novel unseen situations at …

We introduce a value-based RL agent, which we call BBF, that achieves super-human
performance in the Atari 100K benchmark. BBF relies on scaling the neural networks used …

Intelligent agents need to generalize from past experience to achieve goals in complex
environments. World models facilitate such generalization and allow learning behaviors …

The field of meta-learning, or learning-to-learn, has seen a dramatic rise in interest in recent
years. Contrary to conventional approaches to AI where tasks are solved from scratch using …

Meta-reinforcement learning algorithms can enable robots to acquire new skills much more
quickly, by leveraging prior experience to learn how to learn. However, much of the current …

Constructing agents with planning capabilities has long been one of the main challenges in
the pursuit of artificial intelligence. Tree-based planning methods have enjoyed huge …