Pretraining on noisy, internet-scale datasets has been heavily studied a...
An important challenge in reinforcement learning is training agents that...
The promise of reinforcement learning is to solve complex sequential dec...
Quality-Diversity (QD) algorithms, and MAP-Elites (ME) in particular, ha...
This work presents an exploration and imitation-learning-based agent cap...
The structure and performance of neural networks are intimately connecte...
A grand challenge in reinforcement learning is intelligent exploration,
...
An important challenge in reinforcement learning, including evolutionary...
Natural evolution has produced a tremendous diversity of functional
orga...
Hierarchical organization -- the recursive composition of sub-modules --...