Edward Hughes
Scholar

Edward Hughes

Google Scholar ID: 3tj5358AAAAJ
Staff Research Scientist, DeepMind
reinforcement learningmulti-agent systemsopen-endedness
Citations & Impact
All-time
Citations
4,786
 
H-index
29
 
i10-index
34
 
Publications
20
 
Co-authors
53
list available
Resume (English only)
Academic Achievements
  • Publications available on Google Scholar; Visiting Fellow at the London School of Economics, Advisor to the Cooperative AI Foundation, and PIBBSS mentor.
Research Experience
  • Former Staff Research Scientist and Engineer at Google DeepMind, where he co-led the AI Scientist team, worked on the Genie 1 world model, supervised groundbreaking results on cultural evolution in LLMs, and contributed to the AlphaProof system.
Education
  • PhD in Theoretical Physics from Queen Mary University of London, focusing on applications of string theory to particle scattering.
Background
  • A scientific leader in the field of AI with a particular focus on open-ended knowledge discovery. Has made pioneering contributions to fast adaptation in reinforcement learning, the paradigm of Cooperative AI, and ad-hoc collaboration. Draws inspiration from ethics, cultural evolution, social psychology, economics, organisational design, and meta-learning.
Miscellany
  • Cofounder-conductor of the Godwine Choir, with recordings on EM Records and tours in the UK and abroad; supporter of contemporary composers; started his musical career as a choral scholar under Edward Wickham, principal oboist of the Cambridge University Music Society orchestra, and later directed the University of Cambridge Philharmonic Orchestra; former Director of Music at Hornsey Parish Church, London, and Holy Trinity, Hereford; studied conducting with Paul Spicer and Patrick Allies, and voice with Robert Rice; latest discs 'Dream Tryst' and 'Alpha & Omega' are available on Amazon.