87 citations · 159 across the 5 of their papers we have counts for
5 papers
On a Conjecture Regarding the Adam Optimizer
Mohamed Akrout, Douglas Tweed
Why does the Adam optimizer work so well in deep-learning applications? Adam's originators, Kingma and Ba, presented a mathematical argument that was meant to help explain its succ…
Deep Learning without Weight Transport
Mohamed Akrout, Collin Wilson, Peter C. Humphreys +2
Current algorithms for deep learning probably cannot run in the brain because they rely on weight transport, where forward-path neurons transmit their synaptic weights to a feedbac…
Costate-focused models for reinforcement learning
Bita Behrouzi, Xuefei Liu, Douglas Tweed
Many recent algorithms for reinforcement learning are model-free and founded on the Bellman equation. Here we present a method founded on the costate equation and models of the sta…
Random feedback weights support learning in deep neural networks
Timothy P. Lillicrap, Daniel Cownden, Douglas B. Tweed +1
The brain processes information through many layers of neurons. This deep architecture is representationally powerful, but it complicates learning by making it hard to identify the…
Simpler near-optimal controllers through direct supervision
Douglas Tweed
The method of generalized Hamilton-Jacobi-Bellman equations (GHJB) is a powerful way of creating near-optimal controllers by learning. It is based on the fact that if we have a fee…