400 citations · 1.3k across the 51 of their papers we have counts for
21 papers · 1 filter
A Compositional Framework for Open-ended Intelligence
Ida Momennejad, Roberta Raileanu
Open-ended intelligence is the capacity to adapt to novel problems and environments that are substantially different from those in training. A mathematics of open-ended intelligenc…
DiscoGen: Procedural Generation of Algorithm Discovery Tasks in Machine Learning
Alexander D. Goldie, Zilin Wang, Adrian Hayler +17
Automating the development of machine learning algorithms has the potential to unlock new breakthroughs. However, our ability to improve and evaluate algorithm discovery systems ha…
Teaching Large Language Models to Reason with Reinforcement Learning
Alex Havrilla, Yuqing Du, Sharath Chandra Raparthy +6
Reinforcement Learning from Human Feedback (\textbf{RLHF}) has emerged as a dominant approach for aligning LLM outputs with human preferences. Inspired by the success of RLHF, we s…
Generalization to New Sequential Decision Making Tasks with In-Context Learning
Sharath Chandra Raparthy, Eric Hambro, Robert Kirk +2
Training autonomous agents that can learn new tasks from only a handful of demonstrations is a long-standing problem in machine learning. Recently, transformers have been shown to…
The Generalization Gap in Offline Reinforcement Learning
Ishita Mediratta, Qingfei You, Minqi Jiang +1
Despite recent progress in offline learning, these methods are still trained and tested on the same environment. In this paper, we compare the generalization abilities of widely us…
Understanding the Effects of RLHF on LLM Generalisation and Diversity
Robert Kirk, Ishita Mediratta, Christoforos Nalmpantis +4
Large language models (LLMs) fine-tuned with reinforcement learning from human feedback (RLHF) have been used in some of the most widely deployed AI models to date, such as OpenAI'…
On the Importance of Exploration for Generalization in Reinforcement Learning
Yiding Jiang, J. Zico Kolter, Roberta Raileanu
Existing approaches for improving generalization in deep reinforcement learning (RL) have mostly focused on representation learning, neglecting RL-specific aspects such as explorat…
Hyperparameters in Reinforcement Learning and How To Tune Them
Theresa Eimer, Marius Lindauer, Roberta Raileanu
In order to improve reproducibility, deep reinforcement learning (RL) has been adopting better scientific practices such as standardized evaluation metrics and reporting. However,…
MAESTRO: Open-Ended Environment Design for Multi-Agent Reinforcement Learning
Mikayel Samvelyan, Akbir Khan, Michael Dennis +5
Open-ended learning methods that automatically generate a curriculum of increasingly challenging tasks serve as a promising avenue toward generally capable reinforcement learning a…
Building a Subspace of Policies for Scalable Continual Learning
Jean-Baptiste Gaya, Thang Doan, Lucas Caccia +3
The ability to continuously acquire new knowledge and skills is crucial for autonomous agents. Existing methods are typically based on either fixed-size models that struggle to lea…
Dungeons and Data: A Large-Scale NetHack Dataset
Eric Hambro, Roberta Raileanu, Danielle Rothermel +4
Recent breakthroughs in the development of agents to solve challenging sequential decision making problems such as Go, StarCraft, or DOTA, have relied on both simulated environment…
Exploration via Elliptical Episodic Bonuses
Mikael Henaff, Roberta Raileanu, Minqi Jiang +1
In recent years, a number of reinforcement learning (RL) methods have been proposed to explore complex environments which differ across episodes. In this work, we show that the eff…
Insights From the NeurIPS 2021 NetHack Challenge
Eric Hambro, Sharada Mohanty, Dmitrii Babaev +26
In this report, we summarize the takeaways from the first NeurIPS 2021 NetHack Challenge. Participants were tasked with developing a program or agent that can win (i.e., 'ascend' i…
Improving Intrinsic Exploration with Language Abstractions
Jesse Mu, Victor Zhong, Roberta Raileanu +4
Reinforcement learning (RL) agents are particularly hard to train when rewards are sparse. One common solution is to use intrinsic rewards to encourage agents to explore their envi…
Open-Ended Learning Leads to Generally Capable Agents
Open Ended Learning Team, Adam Stooke, Anuj Mahajan +15
In this work we create agents that can perform well beyond a single, individual task, that exhibit much wider generalisation of behaviour to a massive, rich space of challenges. We…
Decoupling Value and Policy for Generalization in Reinforcement Learning
Roberta Raileanu, Rob Fergus
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…
Fast Adaptation via Policy-Dynamics Value Functions
Roberta Raileanu, Max Goldstein, Arthur Szlam +1
Standard RL algorithms assume fixed environment dynamics and require a significant amount of interaction to adapt to new environments. We introduce Policy-Dynamics Value Functions…
The NetHack Learning Environment
Heinrich Küttler, Nantas Nardelli, Alexander H. Miller +4
Progress in Reinforcement Learning (RL) algorithms goes hand-in-hand with the development of challenging environments that test the limits of current methods. While existing RL env…
Automatic Data Augmentation for Generalization in Deep Reinforcement Learning
Roberta Raileanu, Max Goldstein, Denis Yarats +2
Deep reinforcement learning (RL) agents often fail to generalize to unseen scenarios, even when they are trained on many instances of semantically similar environments. Data augmen…
Learning with AMIGo: Adversarially Motivated Intrinsic Goals
Andres Campero, Roberta Raileanu, Heinrich Küttler +3
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…
RIDE: Rewarding Impact-Driven Exploration for Procedurally-Generated Environments
Roberta Raileanu, Tim Rocktäschel
Exploration in sparse reward environments remains one of the key challenges of model-free reinforcement learning. Instead of solely relying on extrinsic rewards provided by the env…