🧑‍🔬 Shane Legg

🏛️ Affiliations

🏢Google DeepMindLondon · United Kingdom2011–2026 · 17 paperscurrent🔬Dalle Molle Institute for Artificial Intelligence ResearchLugano · Switzerland2004–2023 · 18 papers🏢GoogleMountain View · United States2018–2021 · 7 papers🎓University of AucklandAuckland · New Zealand1995–2000 · 3 papers🎓University of Applied Sciences and Arts of Southern SwitzerlandManno · Switzerland2023 · 2 papers🎓Università della Svizzera italianaLugano · Switzerland2011–2023 · 2 papers

From the bylines of their papers (OpenAlex, 2026-10-14), ORCID and Wikidata; institutions on papers are matched by machine. 6 more institutions on single bylines are not listed.

🧭 Fields, topics and concepts

Of its works in our library, with the number of works on each.

📅 Works by year

2020s

2010s

📄 Papers

The most cited first (OpenAlex and our library).

1. Massively Parallel Methods for Deep Reinforcement Learning (2015 · 427 citations)
   DQN, the deep Q-learning agent that learns Atari 2600 games from raw pixels, had only been run on a single machine.
2. Scalable agent alignment via reward modeling: a research direction (2018 · 227 citations)
   Reinforcement learning thrives when a task has an explicit score, as in Atari or Go. Most practical jobs, such as handling email, writing software or managing a business, offer no built-in reward, and the user's goal exists only implicitly. The authors call this the agent alignment problem: making an agent act as its…
3. Neural Networks and the Chomsky Hierarchy (2022 · 112 citations)
   The authors investigate whether the theory of computation, specifically the Chomsky hierarchy, can predict where the ability of neural networks to generalize out of distribution ends.

All works

8 works of theirs in our library, the most cited first; 3 with a page here, the others link their original until we summarise them.

1. IMPALA: Scalable Distributed Deep-RL with Importance Weighted Actor-Learner Architectures ↗arXiv:1802.01561 · no summary yetLasse Espeholt et al.2018 · 710 citations
2. Noisy Networks for Exploration ↗arXiv (Cornell University) · no summary yetMeire Fortunato et al.2017 · 570 citations
3. Massively Parallel Methods for Deep Reinforcement LearningarXiv:1507.04296Arun Nair, Praveen Srinivasan, Sam Blackwell, et al.2015 · 427 citations
4. Scalable agent alignment via reward modeling: a research directionarXiv:1811.07871Jan Leike, David Krueger, Tom Everitt, et al.2018 · 227 citations
5. Neural Networks and the Chomsky HierarchyarXiv:2207.02098Grégoire Delétang, Anian Ruoss, Jordi Grau-Moya, et al.2022 · 112 citations
6. The Hydra Effect: Emergent Self-repair in Language Model Computations ↗arXiv:2307.15771 · no summary yetThomas McGrath et al.2023 · 49 citations
7. Chain of Thought Monitorability: A New and Fragile Opportunity for AI Safety ↗arXiv (Cornell University) · no summary yetTomek Korbak et al.2025 · 35 citations
8. Universal Intelligence: A Definition of Machine Intelligence ↗arXiv:0712.3329 · Minds & Machines, 17:4 (2007) pages 391-444 · no summary yetShane Legg, Marcus Hutter2007 · 23 citations

🤝 Co-authors here

12 of 14 co-authors with a page, the most papers together first.

• Anian Ruoss
• Charles Beattie
• Elliot Catt
• Grégoire Delétang
• Jan Leike
• Joel Veness
• Jordi Grau-Moya
• Li Kevin Wenliang
• Marcus Hutter
• Mustafa Suleyman
• Tim Genewein
• Tom Everitt

🔗 Cited by

Papers of our library cite their works 571 times (of the 149,439 papers whose bibliographies we read); 102 citing papers with a page here →.

Updated 2 November 2026

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/authors/shane-legg/