Warren E. Dixon

Computational Theory and Mathematics · 2003–2013

A researcher in the Science corpus: 2 papers; trusts 7 researchers and is trusted by 4 researchers here, as inferred from the citation graph.

Subfields

Whom Warren E. Dixon trusts overall

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

  1. Frank L. Lewis
  2. Richard S. Sutton
  3. Kenji Doya
  4. Dimitri P. Bertsekas
  5. George V. Cybenko
  6. John J. Hopfield
  7. Andrew Barto

Who trusts Warren E. Dixon overall

  1. Zhiyong Chen
  2. Steven M. LaValle

Papers from Warren E. Dixon's perspective

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

  1. Reinforcement Learning: An Introduction
  2. Approximation by superpositions of a sigmoidal function
  3. Handbook of Intelligent Control: Neural, Fuzzy, and Adaptive Approaches
  4. A Menu of Designs for Reinforcement Learning Over Time
  5. Neuro-Dynamic Programming
  6. Neurons with graded response have collective computational properties like those of two-state neurons.
  7. Neuronlike adaptive elements that can solve difficult learning control problems
  8. Data-driven inversion-based control of nonlinear systems
  9. Online actor critic algorithm to solve the continuous-time infinite horizon optimal control problem (2009 International Joint Conference on Neural Networks)
  10. Handbook of Learning and Approximate Dynamic Programming (IEEE Press Series on Computational Intelligence)

Their papers