Abstract:Evolutionary neural architecture search (ENAS) aims to automate the architecture design of deep neural networks (DNNs). In recent years, various ENAS algorithms have been proposed, and their effectiveness has been demonstrated. In practice, most ENAS methods based on genetic algorithms (GAs) use fixed-length encoding strategies because the generated chromosomes can be directly processed by the standard genetic operators (especially the crossover operator). However, the performance of existing ENAS methods with fixed-length encoding strategies can also be improved because the optimal depth is regarded as a known priori. Although variable-length encoding strategies may alleviate this issue, the standard genetic operators are replaced by the developed operators. In this paper, we proposed a framework to bridge this gap and to improve the performance of existing ENAS methods based on GAs. First, the fixed-length chromosomes were transformed into variable-length chromosomes with the encoding rules of the original ENAS methods. Second, an encoder was proposed to encode variable-length chromosomes into fixed-length representations that can be efficiently processed by standard genetic operators. Third, a decoder cotrained with the encoder was adopted to decode those processed high-dimensional representations which cannot directly describe architectures into original chromosomal forms. Overall, the performances of existing ENAS methods with fixed-length encoding strategies and variable-length encoding strategies have both improved by the proposed framework, and the effectiveness of the framework was justified through experimental results. Moreover, ablation experiments were performed and the results showed that the proposed framework does not negatively affect the original ENAS methods.

Sampled Training and Node Inheritance for Fast Evolutionary Neural Architecture Search

Efficient Evolutionary Search of Attention Convolutional Networks Via Sampled Training and Node Inheritance

An Evolutionary Neural Architecture Search Method based on Performance Prediction and Weight Inheritance

Asynchronous Evolution of Deep Neural Network Architectures

Evolutionary Architecture Search Via Adaptive Parameter Control and Gene Potential Contribution

A Hierarchical Evolution of Neural Architecture Search Method Based on State Transition Algorithm

Evolutionary Neural Architecture Search Based on Evaluation Correction and Functional Units

A Fast Evolutionary Knowledge Transfer Search for Multiscale Deep Neural Architecture

Neural Architecture Search Based on Evolutionary Algorithms with Fitness Approximation

Efficient Self-Learning Evolutionary Neural Architecture Search

A Survey of Advances in Evolutionary Neural Architecture Search

Surrogate-assisted Evolutionary Neural Architecture Search with Network Embedding

Evolutionary Neural Network Architecture Search

Automated Design of CNN Architecture Based on Efficient Evolutionary Search

EGNAS: Efficient Graph Neural Architecture Search Through Evolutionary Algorithm

Multiple Population Alternate Evolution Neural Architecture Search

Benchmarking Analysis of Evolutionary Neural Architecture Search

EENA: Efficient Evolution of Neural Architecture.

Bridge the Gap Between Fixed-Length and Variable-Length Evolutionary Neural Architecture Search Algorithms

Accelerating Evolutionary Neural Architecture Search Via Multi-Fidelity Evaluation

Neural Architecture Search using Progressive Evolution