Correspondence between neuroevolution and gradient descent
arXiv:2008.06643 · doi:10.1038/s41467-021-26568-2
Abstract
We show analytically that training a neural network by conditioned stochastic mutation or neuroevolution of its weights is equivalent, in the limit of small mutations, to gradient descent on the loss function in the presence of Gaussian white noise. Averaged over independent realizations of the learning process, neuroevolution is equivalent to gradient descent on the loss function. We use numerical simulation to show that this correspondence can be observed for finite mutations,for shallow and deep neural networks. Our results provide a connection between two families of neural-network training methods that are usually considered to be fundamentally different.
References in corpus (14)
- Deep Learning in Neural Networks: An Overview
- Playing Atari with Deep Reinforcement Learning
- On the difficulty of training Recurrent Neural Networks
- Evolution Strategies as a Scalable Alternative to Reinforcement Learning
- Deep Neuroevolution: Genetic Algorithms Are a Competitive Alternative for Training Deep Neural Networks for Reinforcement Learning
- Reversible self-assembly of patchy particles into monodisperse icosahedral clusters
- Avoiding unphysical kinetic traps in Monte Carlo simulations of strongly attractive particles
- Rejection-free Geometric Cluster Algorithm for Complex Fluids
- Learning to grow: control of material self-assembly using evolutionary reinforcement learning
- Revisiting the slow dynamics of a silica melt using Monte Carlo simulations
- On the Relationship Between the OpenAI Evolution Strategy and Stochastic Gradient Descent
- Approximating the dynamical evolution of systems of strongly-interacting overdamped particles
- Guided evolutionary strategies: Augmenting random search with surrogate gradients
- Evolution Strategies Converges to Finite Differences