Stuart Dreyfus

Professor Emeritus

Ph.D. Harvard University, 1964
Applied Mathematics

4125 Etcheverry Hall
(510) 642-4926
E-mail: dreyfus@ieor.berkeley.edu

"Expertise is pattern discrimination and association based on experience. It is intuitive. There is no evidence you can reduce it to rules and theory. Hence, Artificial Intelligence probably can't be produced using rules and principles. That's not what intelligence is."


Research
  • Neural Networks
  • Dynamic Programming
  • Limits of Operations Research Modeling
  • Cognitive Ergonomics

Recent Publications
  • Thinking slow about thinking fast.
  • System 0: the overlooked explanation of expert intuition.
  • Beyond expertise: some preliminary thoughts on mastery. Authors: Hubert L. Dreyfus and Stuart E. Dreyfus.
  • A Modern Perspective on Creative Cognition Bulletin of Science, Technology & Society, February 2009 vol. 29 pages 3-8.
  • Neuroscience and Operations Research OR/MS April 2010, Forums
  • Modern Computational Applications of Dynamic Programming Journal of Industrial and Systems Engineering, Vol. 4, No. 3, pp 152-155 Fall 2010
  • Errata for Eiji Mizutani and Stuart E. Dreyfus.``Second-order stagewise backpropagation for Hessian-matrix analyses and investigation of negative curvature.'' Neural Networks, pages 193--203, Vol.21, 2008.
  • Errata
  • Eiji Mizutani and Stuart E. Dreyfus. ``On using discretized Cohen-Grossberg node dynamics for model-free actor-critic neural learning in non-Markovian domains,'' In Proc. of the 5th IEEE International Symposium on Computational Intelligence in Robotics and Automation (CIRA 2003), Vol.1, pp. 1--6, Kobe JAPAN, July 2003. cira03
  • Eiji Mizutani and Stuart E. Dreyfus. ``On practical use of stagewise second-order backpropagation for multi-stage neural-network learning,'' In Proc. of the 2007 International Joint Conference on Neural Networks, Orlando FL., August 2007. ESijcnn2007
  • Eiji Mizutani and Stuart Dreyfus. ``On derivation of stagewise second-order backpropagation by invariant imbedding for multi-stage neural-network learning.'' In Proc. of the 2006 International Joint Conf. on Neural Networks, part of the IEEE World Congress on Computational Intelligence, Vancouver, CANADA, July 2006. ESCCI06
  • Eiji Mizutani, Stuart E. Dreyfus, and James W. Demmel. ``Second-order backpropagation algorithms for a stagewise-partitioned separable Hessian matrix.'' In Proceedings of the International Joint Conference on Neural Networks, Montreal, Canada, 2005. IJCNN05.pdf
  • E. Mizutani and S. E. Dreyfus "Stagewise Newton, differential dynamic programming, neighboring optimum control for neural-network learning". In Proceedings of the the 2005 American Control Conference, Portland Oregon, June 8-10, 2005. ACC05.pdf
  • Eiji Mizutani and Stuart Dreyfus. ``Two Stochastic Dynamic Programming Problems by Model-Free Actor-Critic Recurrent Network Learning in Non-Markovian Settings.'' In Proceedings of the IEEE-INNS International Joint Conference on Neural Networks, Budapest, Hungary, July 25-29, 2004. ijcnn04_1.pdf
  • S. E. Dreyfus "Totally Model-Free Learned Skillful Coping ." Bulletin of Science, Technology & Society, Vol. 24, No.3, June 2004, 182-187.
  • E. Mizutani and S. E. Dreyfus "Matlab code for hidden-node teaching". hidteach.m
  • E. Mizutani and S. E. Dreyfus "On complexity analysis of supervised MLP-learning for algorithmic comparisons". To appear in Proceedings of: INNS-IEEE International Joint Conference on Neural Networks Washington D.C. July 14-19, 2001 IJCNN2001.pdf
  • E. Mizutani, S. E. Dreyfus, and K. Nishio "On derivation of MLP backpropagation from the Kelley-Bryson optimal-control gradient formula and its application".Proceedings of the IEEE International Joint Conference on Neural Networks (IJCNN 2000), Como Italy, July 2000 ijcnn2k.pdf
  • E. Mizutani, S. E. Dreyfus, and J.-S. Roger Jang "On dynamic programming-like recursive gradient formula for alleviating MLP hidden-node saturation in the parity problem".Proceedings of the 8th Bellman Continuum International Workshop, Hsinchu, TAIWAN, December 2000 (To appear) bellman00.pdf
  • E. Mizutani and S. E Dreyfus "Totally Model-Free Reinforcement Learning by Actor-Critic Elman Networks in Non-Markovian Domains".Proceedings of the IEEE World Congress on Computational Intelligence (Wcci'98), Alaska USA, May 1998 pp.2016--2021 Wcci98.pdf
  • "Artificial Neural Networks, Back Propagation and the Kelley-Bryson Gradient Procedure,"J. Guidance, Control and Dynamics, 1990.
  • "How Reliable are Computer Models of Socioeconomic Systems?,"ACM Conference on Critical Issues, 1990.
  • "Making a Mind Versus Modeling the Brain: Artificial Intelligence Back at a Branchpoint" (with H. Dreyfus),Daedulus, Winter 1988
  • Mind Over Machine (with H. Dreyfus), Free Press, 1986

Recent Ph.D. Theses Supervised

  • "Self-Organizing Neural Networks in Decision Making Application," Laura Burke
  • "Improved Learning Algorithm of Neural Networks," R. Tsaih
  • "Input Feature Sealing Algorithm for Competitive-Learning-Based Cognitive Modeling with Two Applications," Choon Leem
  • "A Neural Net Knowledge_based System with Instance-Based Rules," Jongtae Rhee