Výsledky vyhledávání - "Tweed, Douglas"

Report

On a Conjecture Regarding the Adam Optimizer

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 success, but Bock and colleagues have since reported that a key piece is m

Externí odkaz: http://arxiv.org/abs/2111.08162

Zobrazit plný text záznamu

Report

Deep Learning without Weight Transport

Autor: Akrout, Mohamed, Wilson, Collin, Humphreys, Peter C., Lillicrap, Timothy, Tweed, Douglas

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 feedback path, in a way that is likely impossible biologically. An algorithm

Externí odkaz: http://arxiv.org/abs/1904.05391

Zobrazit plný text záznamu

Report

Costate-focused models for reinforcement learning

Autor: Behrouzi, Bita, Liu, Xuefei, Tweed, Douglas

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 state dynamics. We use the costate -- the gradient of cost with respect t

Externí odkaz: http://arxiv.org/abs/1711.05817

Zobrazit plný text záznamu

Report

Random feedback weights support learning in deep neural networks

Autor: Lillicrap, Timothy P., Cownden, Daniel, Tweed, Douglas B., Akerman, Colin J.

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 responsible neurons when a mistake is made. In machine learning, the

Externí odkaz: http://arxiv.org/abs/1411.0247

Zobrazit plný text záznamu

Report

Simpler near-optimal controllers through direct supervision

Autor: Tweed, Douglas

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 feedback controller, and we learn to compute the gradient grad-J of its c

Externí odkaz: http://arxiv.org/abs/0908.2859

Zobrazit plný text záznamu