
The UCL Deciding, Acting, and Reasoning with Knowledge (DARK) Lab is a Reinforcement Learning research group at the UCL Centre for Artificial Intelligence. We focus on research in complex open-ended environments that provide a constant stream of novel observations without reliable reward functions, often requiring agents to create their own curricula and to deal with external knowledge, natural language, and hard exploration problems.
News
-
25/09/2023:
Large language models are not zero-shot communicators has been accepted to NeurIPS 2023 (spotlight).
-
25/09/2023:
SMACv2: An Improved Benchmark for Cooperative Multi-Agent Reinforcement Learning has been accepted to NeurIPS 2023.
-
24/01/2023:
MAESTRO: Open-Ended Environment Design for Multi-Agent Reinforcement Learning has been accepted to ICLR 2023.
-
24/01/2023:
Efficient Planning in a Compact Latent Action Space has been accepted to ICLR 2023.
-
09/01/2023:
A Survey of Zero-shot Generalisation in Deep Reinforcement Learning has been published in JAIR.
-
8/11/2022:
Grounding Aleatoric Uncertainty for Unsupervised Environment Design has been accepted to NeurIPS 2022.
-
8/11/2022:
Learning General World Models in a Handful of Reward-Free Deployments has been accepted to NeurIPS 2022.
-
8/11/2022:
Exploration via Elliptical Episodic Bonuses has been accepted to NeurIPS 2022.
Faculty
PhD Students
Alumni
Hannah Teufel
MSc Student (2021)
