Generic Neural Architecture Search via Regression

Most existing neural architecture search (NAS) algorithms are dedicated to and evaluated by the downstream tasks, e.g., image classification in computer vision. However, extensive experiments have shown that, prominent neural architectures, such as ResNet in computer vision and LSTM in natural language processing, are generally good at extracting patterns from the input data and perform well on different downstream tasks. In this paper, we attempt to answer two fundamental questions related to NAS. (1) Is it necessary to use the performance of specific downstream tasks to evaluate and search for good neural architectures? (2) Can we perform NAS effectively and efficiently while being agnostic to the downstream tasks? To answer these questions, we propose a novel and generic NAS framework, termed Generic NAS (GenNAS). GenNAS does not use task-specific labels but instead adopts regression on a set of manually designed synthetic signal bases for architecture evaluation. Such a self-supervised regression task can effectively evaluate the intrinsic power of an architecture to capture and transform the input signal patterns, and allow more sufficient usage of training samples. Extensive experiments across 13 CNN search spaces and one NLP space demonstrate the remarkable efficiency of GenNAS using regression, in terms of both evaluating the neural architectures (quantified by the ranking correlation Spearman's rho between the approximated performances and the downstream task performances) and the convergence speed for training (within a few seconds).

PDF Abstract NeurIPS 2021 PDF NeurIPS 2021 Abstract

Results from the Paper

 Ranked #1 on Neural Architecture Search on NAS-Bench-101 (Spearman Correlation metric)

     Get a GitHub badge
Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Neural Architecture Search NAS-Bench-101 GenNAS Spearman Correlation 0.87 # 1
Neural Architecture Search NAS-Bench-201, CIFAR-10 GenNAS Accuracy (Test) 94.18 # 12
Accuracy (Val) - # 31
Search time (s) 1080 # 5
Neural Architecture Search NAS-Bench-201, CIFAR-100 GenNAS Accuracy (Test) 72.56 # 14
Search time (s) 1080 # 5
Neural Architecture Search NAS-Bench-201, ImageNet-16-120 GenNAS Accuracy (Test) 45.59 # 22
Search time (s) 1080 # 7