Feynman on Artificial Intelligence and Machine Learning
Eric Mjolsness · 2023
Richard Feynman&s;s role in sparking the development of quantum computing is well known, as well as his affinity for calculating machines. The Physics of Computation course began as a collaborative offering between Feynman, Hopfield, and Carver Mead, the pioneer of very large-scale integration computer chip design, who was also interested in neural computation as a new hardware paradigm. Neural networks are usually specified with a combination of algebraic equations with free parameters and illustrated by block diagrams showing how variables are connected to one another in these equations. In the realm of computer algebra for symbolic integration and differential equation solving, there is good progress in approaching these problems by means of deep neural networks, although more remains to be done. In summary, many of Richard Feynman&s;s ideas about the future of neural network-like machine learning systems have come to pass.