Safe Mutations for Deep and Recurrent Neural Networks through Output Gradients

by   Joel Lehman, et al.

While neuroevolution (evolving neural networks) has a successful track record across a variety of domains from reinforcement learning to artificial life, it is rarely applied to large, deep neural networks. A central reason is that while random mutation generally works in low dimensions, a random perturbation of thousands or millions of weights is likely to break existing functionality, providing no learning signal even if some individual weight changes were beneficial. This paper proposes a solution by introducing a family of safe mutation (SM) operators that aim within the mutation operator itself to find a degree of change that does not alter network behavior too much, but still facilitates exploration. Importantly, these SM operators do not require any additional interactions with the environment. The most effective SM variant capitalizes on the intriguing opportunity to scale the degree of mutation of each individual weight according to the sensitivity of the network's outputs to that weight, which requires computing the gradient of outputs with respect to the weights (instead of the gradient of error, as in conventional deep learning). This safe mutation through gradients (SM-G) operator dramatically increases the ability of a simple genetic algorithm-based neuroevolution method to find solutions in high-dimensional domains that require deep and/or recurrent neural networks (which tend to be particularly brittle to mutation), including domains that require processing raw pixels. By improving our ability to evolve deep neural networks, this new safer approach to mutation expands the scope of domains amenable to neuroevolution.


page 12

page 13


Hybridizing PSM and RSM Operator for Solving NP-Complete Problems: Application to Travelling Salesman Problem

In this paper, we present a new mutation operator, Hybrid Mutation (HPRM...

Improving TSP Solutions Using GA with a New Hybrid Mutation Based on Knowledge and Randomness

Genetic algorithm (GA) is an efficient tool for solving optimization pro...

Safety and Robustness in Decision Making: Deep Bayesian Recurrent Neural Networks for Somatic Variant Calling in Cancer

The genomic profile underlying an individual tumor can be highly informa...

DXNN Platform: The Shedding of Biological Inefficiencies

This paper introduces a novel type of memetic algorithm based Topology a...

An Experimental Study of Weight Initialization and Weight Inheritance Effects on Neuroevolution

Weight initialization is critical in being able to successfully train ar...

Infinite-dimensional Folded-in-time Deep Neural Networks

The method recently introduced in arXiv:2011.10115 realizes a deep neura...

Evolving A-Type Artificial Neural Networks

We investigate Turing's notion of an A-type artificial neural network. W...

Please sign up or login with your details

Forgot password? Click here to reset