BossNAS: Exploring Hybrid CNN-transformers with Block-wisely Self-supervised Neural Architecture Search

TL;DR

BossNAS introduces a self-supervised neural architecture search method that effectively explores hybrid CNN-transformer architectures by block-wise training, achieving state-of-the-art accuracy on ImageNet with improved architecture rating correlations.

Contribution

The paper proposes a novel unsupervised NAS approach with ensemble bootstrapping for block-wise training, enabling efficient search in hybrid CNN-transformer spaces.

Findings

Achieves 82.5% ImageNet accuracy with BossNet-T.

Surpasses EfficientNet in accuracy with similar compute.

Outperforms state-of-the-art NAS methods in architecture rating correlation.

Abstract

A myriad of recent breakthroughs in hand-crafted neural architectures for visual recognition have highlighted the urgent need to explore hybrid architectures consisting of diversified building blocks. Meanwhile, neural architecture search methods are surging with an expectation to reduce human efforts. However, whether NAS methods can efficiently and effectively handle diversified search spaces with disparate candidates (e.g. CNNs and transformers) is still an open question. In this work, we present Block-wisely Self-supervised Neural Architecture Search (BossNAS), an unsupervised NAS method that addresses the problem of inaccurate architecture rating caused by large weight-sharing space and biased supervision in previous methods. More specifically, we factorize the search space into blocks and utilize a novel self-supervised training scheme, named ensemble bootstrapping, to train each block separately before searching them as a whole towards the population center. Additionally, we present HyTra search space, a fabric-like hybrid CNN-transformer search space with searchable down-sampling positions. On this challenging search space, our searched model, BossNet-T, achieves up to 82.5% accuracy on ImageNet, surpassing EfficientNet by 2.4% with comparable compute time. Moreover, our method achieves superior architecture rating accuracy with 0.78 and 0.76 Spearman correlation on the canonical MBConv search space with ImageNet and on NATS-Bench size search space with CIFAR-100, respectively, surpassing state-of-the-art NAS methods. Code: https://github.com/changlin31/BossNAS