Paul Werbos: The Evolution of Artificial Neural Networks
The applications of artificial neural networks are legion. Today, Robert J. Marks talks with Dr. Paul Werbos, the man who invented the method used for over four decades to train artificial neural networks. The two discuss Werbos’s mathematical journey, the error backpropagation algorithm, and the slog of “making it” in scientific academic research.
Show Notes
- 01:19 | Introducing Dr. Paul Werbos
- 02:33 | Werbos’s error backpropagation algorithm
- 07:39 | Marvin Minsky and neural networks
- 10:34 | recurrent neural networks and feedback
- 12:34 | Harvard & MIT’s Cambridge Project
- 13:36 | Werbos’s “flash of genius”
- 18:36 | Pushback from the PhD committee
- 22:16 | Dynamic feedback
- 26:01 | Does a version of error backpropagation occur in the human brain?
- 29:24 | Changing the game
Additional Resources
- Dr. Paul Werbos at IEEE.org
- Paul Werbos’s website
- Paul Werbos’s PhD dissertation introducing error backpropagation used today to train artificial neural networks
- Talking Nets: An Oral History of Neural Networks at Amazon.com
- Paul Werbos’s 1972 Proposal to Harvard for Backpropagation and Intelligent Reinforcement System
- Artificial Intelligence in the Age of Neural Networks and Brain Computing on Amazon.com
- NSF Award granted to Andrew Ng and Yann LeCun
- “Mind, Brain and Soul From the Viewpoint of Mathematical Realism” at www.werbos.com
- Computers and Thought on Amazon.com
- John von Neumann, Hungarian-American mathematician, physicist, and computer scientist
- Perceptrons: An Introduction to Computational Geometry by Marvin Minsky and Seymour A. Papert on Amazon.com
- Paul Erdős, Hungarian mathematician
- Alonzo Church, American mathematician
- Alan Turing, British mathematician and logician
- Jean-Baptise Fourier, French mathematician and physicist
- György Buzsáki, Biggs Professor of Neuroscience at New York University School of Medicine
- Barry Richmond, Principal Investigator at NIH
