Graham Kendall

Artificial Intelligence · 2000–2013

A researcher in the Science corpus: 9 papers; trusts 10 researchers and is trusted by 14 researchers here, as inferred from the citation graph.

Subfields

Whom Graham Kendall trusts overall

Their strongest outbound trust, strongest first, inferred from their citations.

  1. Richard S. Sutton
  2. John R. Koza
  3. Andrew Barto
  4. Edmund Kieran Burke
  5. Leslie Pack Kaelbling
  6. Krzysztof Socha
  7. Jonas Mockus
  8. Pierre Hansen
  9. Peter I. Cowling

Who trusts Graham Kendall overall

  1. Edmund Kieran Burke
  2. Rhyd Lewis
  3. Ferrante Neri
  4. Krzysztof Socha
  5. Matthew R. Hyde
  6. Barry McCollum
  7. Carlos Cotta
  8. Tansel Dökeroğlu
  9. J. Branke

Papers from Graham Kendall's perspective

Ordered by how strongly the researchers Graham Kendall trusts endorse each paper. Same corpus, a different order for every researcher.

  1. Reinforcement Learning: A Survey
  2. Genetic Programming: On the Programming of Computers by Means of Natural Selection
  3. Reinforcement learning
  4. The Arcade Learning Environment: An Evaluation Platform for General Agents
  5. In Modern Heuristic Techniques for Combinatorial Problems
  6. Bandit Based Monte-Carlo Planning
  7. The NP-Completeness Column: An Ongoing Guide
  8. Variable neighborhood search (European Journal of Operational Research)
  9. A Hyperheuristic Approach to Scheduling a Sales Summit (International Conference on the Practice and Theory of Automated Timetabling)
  10. The Bayesian approach to global optimization

Their papers