Fact sheet

Class
researcher

Stuart E. Dreyfus is professor emeritus of industrial engineering and operations research at the University of California, Berkeley.[1] At the RAND Corporation he programmed the JOHNNIAC computer and co-authored Applied Dynamic Programming with Richard Bellman; dynamic programming, his central field, is one of the mathematical foundations of the reinforcement learning behind modern AI.[1]

With his younger brother Hubert, a philosopher, he built the Dreyfus model of skill acquisition, first set out in their 1980 report for the Air Force.[2] He later argued, in his account of “System 0,” that expert intuition is the brain’s procedural memory learning by trial, feedback, and reward, the same reinforcement-learning mechanism his own mathematics helped give the machines.[3]

References

1
2
Stuart E. Dreyfus and Hubert L. Dreyfus, A Five-Stage Model of the Mental Activities Involved in Directed Skill Acquisition, ORC 80-2 (Berkeley: Operations Research Center, University of California, Berkeley, 1980).
https://apps.dtic.mil/sti/tr/pdf/ADA084551.pdf
3
Stuart E. Dreyfus, "System 0: The Overlooked Explanation of Expert Intuition," in Handbook of Research Methods on Intuition, ed. Marta Sinclair (Cheltenham: Edward Elgar, 2014).
https://escholarship.org/uc/item/7nk534tm

Keep learning

Dreyfus ModelToolEvery skill runs on a ladder from novice to expert, with a sixth rung, mastery, for those who push further, and where you sit changes how you learn it and how you should be coached. The Dreyfus model is the map: name the rung, name the next one, and the vague feedback that helps no one turns into a plan.