IEEE Citation for Frank Rosenblatt Award, top technical field award for computational intelligence

2022 IEEE FRANK ROSENBLATT AWARD Sponsored by the IEEE Computational Intelligence Society PAUL JOHN WERBOS For development of backpropagation and fundamental contributions to reinforcement learning and time series analysis Among the first researchers to realize the power of bio-inspired learning techniques to train neural networks in real time, Paul John Werbos’ development of backpropagation algorithms provided the backbone of reinforcement and deep learning methods for solving today’s complex tasks. Backpropagation allows training of neural network data online and in real time by using gradients computed backward through the layers of the neural network. His leadership of the Adaptive and Intelligent Systems group at the U.S. National Science Foundation enhanced the ability of countless researchers to contribute to prediction and control of systems ranging from nanorobots to the electric power grid. His work has made possible many advances in areas including electric vehicles and speech, face, and handwriting recognition applications. An IEEE Fellow, Werbos is program director (retired) with the National Science Foundation, Arlington, Virginia, USA. TAGLINE: Backpropagation is the backbone of neural network training for deep learning applications critical to prediction and control of complex systems