Papers › BossNAS: Exploring Hybrid CNN-transformers with Block-wisely Self-supervised Neural...

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

23 Mar 2021ICCV 2021 10arXiv:2103.12424archive 2025-07-28

Changlin Li, Tao Tang, Guangrun Wang, Jiefeng Peng, Bing Wang, Xiaodan Liang, Xiaojun Chang

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

PaperPDFConference PDFCode

In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

Code

changlin31/BossNAS officialmentioned in papermentioned on GitHubpytorch report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

Image ClassificationNeural Architecture SearchOpen-Ended Question Answering

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Image Classification ImageNet BossNet-T1 GFLOPs 15.8 #558 of 1060 Archive leaderboard report
Image Classification ImageNet BossNet-T1 Top 1 Accuracy 82.2% #558 of 1060 Archive leaderboard report
Neural Architecture Search ImageNet BossNet-T1+ Accuracy 82.2 #6 of 135 Archive leaderboard report
Neural Architecture Search ImageNet BossNet-T1+ MACs 10.5G #6 of 135 Archive leaderboard report
Neural Architecture Search ImageNet BossNet-T1+ Top-1 Error Rate 17.8 #6 of 135 Archive leaderboard report
Neural Architecture Search NATS-Bench Size, CIFAR-10 BossNAS Acc. (test) 93.29 #4 of 5 Archive leaderboard report
Neural Architecture Search NATS-Bench Size, CIFAR-10 BossNAS Kendall's Tau 0.53 #4 of 5 Archive leaderboard report
Neural Architecture Search NATS-Bench Size, CIFAR-10 BossNAS Pearson R 0.72 #4 of 5 Archive leaderboard report
Neural Architecture Search NATS-Bench Size, CIFAR-10 BossNAS Spearman's Rho 0.73 #4 of 5 Archive leaderboard report
Neural Architecture Search NATS-Bench Size, CIFAR-100 BossNAS Acc. (test) 70.86 #4 of 5 Archive leaderboard report
Neural Architecture Search NATS-Bench Size, CIFAR-100 BossNAS Kendall's Tau 0.59 #4 of 5 Archive leaderboard report
Neural Architecture Search NATS-Bench Size, CIFAR-100 BossNAS Pearson R 0.79 #4 of 5 Archive leaderboard report
Neural Architecture Search NATS-Bench Size, CIFAR-100 BossNAS Spearman's Rho 0.76 #4 of 5 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Methods

1x1 ConvolutionAverage PoolingBatch NormalizationConvolutionDense ConnectionsDepthwise ConvolutionDepthwise Separable ConvolutionDropoutEfficientNetInverted Residual BlockPointwise ConvolutionRMSPropReLUSigmoid ActivationSqueeze-and-Excitation Block

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections