Deep Neural Architecture Search with Deep Graph Bayesian Optimization

TL;DR

This paper introduces a novel graph neural network-based Bayesian optimization method for neural architecture search, automatically learning features from architectures to improve search efficiency and performance.

Contribution

It proposes a graph neural network surrogate within Bayesian optimization for neural architecture search, eliminating the need for handcrafted similarity metrics.

Findings

Significantly outperforms existing methods on benchmark tasks

Automatically extracts features from neural architectures

Enhances the efficiency of neural architecture search

Abstract

Bayesian optimization (BO) is an effective method of finding the global optima of black-box functions. Recently BO has been applied to neural architecture search and shows better performance than pure evolutionary strategies. All these methods adopt Gaussian processes (GPs) as surrogate function, with the handcraft similarity metrics as input. In this work, we propose a Bayesian graph neural network as a new surrogate, which can automatically extract features from deep neural architectures, and use such learned features to fit and characterize black-box objectives and their uncertainty. Based on the new surrogate, we then develop a graph Bayesian optimization framework to address the challenging task of deep neural architecture search. Experiment results show our method significantly outperforms the comparative methods on benchmark tasks.