A Self-adaptive Neuroevolution Approach to Constructing Deep Neural Network Architectures Across Different Types

TL;DR

This paper introduces SANE, a self-adaptive neuroevolution method that automatically constructs lightweight, effective DNN architectures across various types by self-adapting the search space and evolution process.

Contribution

The paper presents a novel self-adaptive neuroevolution approach that dynamically adjusts the search space and evolution strategy for diverse DNN architectures.

Findings

Generated DNNs are smaller with comparable performance to existing architectures.

SANE effectively adapts to different DNN types like CNN, GAN, and LSTM.

The method improves search efficiency through self-adaptation and speciation.

Abstract

Neuroevolution has greatly promoted Deep Neural Network (DNN) architecture design and its applications, while there is a lack of methods available across different DNN types concerning both their scale and performance. In this study, we propose a self-adaptive neuroevolution (SANE) approach to automatically construct various lightweight DNN architectures for different tasks. One of the key settings in SANE is the search space defined by cells and organs self-adapted to different DNN types. Based on this search space, a constructive evolution strategy with uniform evolution settings and operations is designed to grow DNN architectures gradually. SANE is able to self-adaptively adjust evolution exploration and exploitation to improve search efficiency. Moreover, a speciation scheme is developed to protect evolution from early convergence by restricting selection competition within species. To evaluate SANE, we carry out neuroevolution experiments to generate different DNN architectures including convolutional neural network, generative adversarial network and long short-term memory. The results illustrate that the obtained DNN architectures could have smaller scale with similar performance compared to existing DNN architectures. Our proposed SANE provides an efficient approach to self-adaptively search DNN architectures across different types.