Ronald J. Williams

Current role not listed

Williams was a neural-network pioneer who co-authored the landmark 1986 backpropagation paper and later introduced REINFORCE, the first policy-gradient algorithm. He also helped develop teacher forcing for training recurrent networks.