Author:
Kubo Yoshimasa,Chalmers Eric,Luczak Artur
Abstract
AbstractBackpropagation has been used to train neural networks for many years, allowing them to solve a wide variety of tasks like image classification, speech recognition, and reinforcement learning tasks. But the biological plausibility of backpropagation as a mechanism of neural learning has been questioned. Equilibrium Propagation (EP) has been proposed as a more biologically plausible alternative and achieves comparable accuracy on the CIFAR-10 image classification task. This study proposes the first EP-based reinforcement learning architecture: an actor-critic architecture with the actor network trained by EP. We show that this model can solve the basic control tasks often used as benchmarks for BP-based models. Interestingly, our trained model demonstrates more consistent high-reward behavior than a comparable model trained exclusively by backpropagation.
Publisher
Cold Spring Harbor Laboratory
Reference40 articles.
1. Learning representations by back-propagating errors
2. Mastering the game of Go with deep neural networks and tree search
3. Human-level control through deep reinforcement learning
4. Explaining and harnessing adversarial examples;arXiv preprint,2014
5. Simple statistical gradient-following algorithms for connectionist reinforcement learning;Machine learning,1992