Collaborating with humans without human data
Collaborating with humans requires rapidly adapting to their individual strengths,
weaknesses, and preferences. Unfortunately, most standard multi-agent reinforcement …
weaknesses, and preferences. Unfortunately, most standard multi-agent reinforcement …
Emergent complexity and zero-shot transfer via unsupervised environment design
A wide range of reinforcement learning (RL) problems---including robustness, transfer
learning, unsupervised RL, and emergent complexity---require specifying a distribution of …
learning, unsupervised RL, and emergent complexity---require specifying a distribution of …
Contrastive behavioral similarity embeddings for generalization in reinforcement learning
Reinforcement learning methods trained on few environments rarely learn policies that
generalize to unseen environments. To improve generalization, we incorporate the inherent …
generalize to unseen environments. To improve generalization, we incorporate the inherent …
The nethack learning environment
Abstract Progress in Reinforcement Learning (RL) algorithms goes hand-in-hand with the
development of challenging environments that test the limits of current methods. While …
development of challenging environments that test the limits of current methods. While …
Decoupling value and policy for generalization in reinforcement learning
R Raileanu, R Fergus - International Conference on …, 2021 - proceedings.mlr.press
Standard deep reinforcement learning algorithms use a shared representation for the policy
and value function, especially when training directly from images. However, we argue that …
and value function, especially when training directly from images. However, we argue that …
Learning with amigo: Adversarially motivated intrinsic goals
A Campero, R Raileanu, H Küttler… - arXiv preprint arXiv …, 2020 - arxiv.org
A key challenge for reinforcement learning (RL) consists of learning in environments with
sparse extrinsic rewards. In contrast to current RL methods, humans are able to learn new …
sparse extrinsic rewards. In contrast to current RL methods, humans are able to learn new …
Neuroevolution of self-interpretable agents
Inattentional blindness is the psychological phenomenon that causes one to miss things in
plain sight. It is a consequence of the selective attention in perception that lets us remain …
plain sight. It is a consequence of the selective attention in perception that lets us remain …
Automatic data augmentation for generalization in reinforcement learning
R Raileanu, M Goldstein, D Yarats… - Advances in …, 2021 - proceedings.neurips.cc
Deep reinforcement learning (RL) agents often fail to generalize beyond their training
environments. To alleviate this problem, recent work has proposed the use of data …
environments. To alleviate this problem, recent work has proposed the use of data …
On the importance of exploration for generalization in reinforcement learning
Existing approaches for improving generalization in deep reinforcement learning (RL) have
mostly focused on representation learning, neglecting RL-specific aspects such as …
mostly focused on representation learning, neglecting RL-specific aspects such as …
Warmth and competence in human-agent cooperation
Interaction and cooperation with humans are overarching aspirations of artificial intelligence
research. Recent studies demonstrate that AI agents trained with deep reinforcement …
research. Recent studies demonstrate that AI agents trained with deep reinforcement …