AI Paper Atlas

Research map / Label-efficient and robust learning / Self-supervised and contrastive

Synergistic Self-Supervised and Quantization Learning

Yun-Hao Cao, Peiqin Sun, Yechang Huang, Jianxin Wu, Shuchang Zhou

ECCV 2022 · Label-efficient and robust learning › Self-supervised and contrastive · approach: Data-efficient and robust self-supervision · 11 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
quantize only the target branch (no gradient to quantized branch) (ablation)SimSiam on CIFAR-10 · CIFAR-10 linear eval ResNet-34, averageavg accuracy: 79.2 -> 78.0Hurt
also quantize the prediction branch and update it with gradients (ablation)SSQL · CIFAR-10 linear eval ResNet-34, averageavg accuracy: 78.0 -> 84.5Improved
randomly select bit-widths instead of a single 4-bit setting (ablation)SSQL · CIFAR-10 linear eval ResNet-34, averageavg accuracy: 33.4 -> 83.9Improved
randomly select bit-widths instead of a single 6-bit setting (ablation)SSQL · CIFAR-10 linear eval ResNet-34, averageavg accuracy: 79.8 -> 83.9Improved
use narrower low-bit range 2~8/4~8 instead of 4~16 bits (ablation)SSQL · CIFAR-10 linear eval ResNet-34, 2w4aaccuracy: 65.2 -> 81.5Mixed
add auxiliary SimSiam loss (ablation)SSQL · CIFAR-10 linear eval ResNet-34, averageavg accuracy: 86.3 -> 88.3Improved
also quantize the target branch with 2~8 bit range (ablation)SSQL · CIFAR-10 linear eval ResNet-34, averageavg accuracy: 86.3 -> 85.8Hurt
use SSQL pretraining instead of SimSiamResNet-18 · CIFAR-10 linear eval (3w3a)accuracy: 66.0 -> 85.6Improved
use SSQL pretraining instead of SimSiamResNet-18 · ImageNet linear eval (FP)accuracy: 55.0 -> 57.6Improved
use SSQL pretraining instead of SimSiamResNet-50 · ImageNet linear eval (4w4a)accuracy: 52.4 -> 63.0Improved
add SSQL to BYOLResNet-18 · CIFAR-10 linear eval (2w4a)accuracy: 63.3 -> 85.7Improved
add SSQL to MoCov3ViT-Small · CIFAR-10 linear eval (4w4a)accuracy: 82.0 -> 86.9Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy