Recombination of Artificial Neural Networks
This is an incremental improvement to population-based training methods for neural network optimization.
The paper tackles hyperparameter optimization for neural networks by proposing a genetic algorithm that combines gradient-based training with population-based methods, using three-parent reproduction to separate parameters from hyperparameters. The result shows improved final accuracy and faster convergence across multiple network architectures.
We propose a genetic algorithm (GA) for hyperparameter optimization of artificial neural networks which includes chromosomal crossover as well as a decoupling of parameters (i.e., weights and biases) from hyperparameters (e.g., learning rate, weight decay, and dropout) during sexual reproduction. Children are produced from three parents; two contributing hyperparameters and one contributing the parameters. Our version of population-based training (PBT) combines traditional gradient-based approaches such as stochastic gradient descent (SGD) with our GA to optimize both parameters and hyperparameters across SGD epochs. Our improvements over traditional PBT provide an increased speed of adaptation and a greater ability to shed deleterious genes from the population. Our methods improve final accuracy as well as time to fixed accuracy on a wide range of deep neural network architectures including convolutional neural networks, recurrent neural networks, dense neural networks, and capsule networks.