AI Paper Atlas

Research map / Label-efficient and robust learning / Knowledge distillation

Dataset Distillation: A Comprehensive Review

Ruonan Yu, Songhua Liu, Xinchao Wang

TPAMI 2023 · Label-efficient and robust learning › Knowledge distillation · approach: Self and data-free distillation · 210 citations (OpenAlex)

Read the paper

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
use trajectory/kernel-based matching (FRePo) instead of gradient matching (DC)dataset distillation · CIFAR-100 (10 Img/Cls)test accuracy (%): 25.4 ± 0.3 -> 41.3 ± 0.2Improved
use multi-step parameter matching (MTT) instead of distribution matching (DM)dataset distillation · CIFAR-10 (1 Img/Cls)test accuracy (%): 26.5 ± 0.4 -> 46.3 ± 0.8Improved
use multi-step parameter matching (MTT) instead of single-step gradient matching with augmentation (DSA)dataset distillation · Tiny-ImageNet (10 Img/Cls)test accuracy (%): 16.3 ± 0.2 -> 23.2 ± 0.2Improved
add momentum to inner-loop parameter update (DD vs DC)dataset distillation · CIFAR-10 (10 Img/Cls)test accuracy (%): 44.9 ± 0.5 -> 60.2 ± 0.4Improved
add momentum in inner loopDD · MNIST/CIFARtest accuracy (%)Improved
evaluate DD-distilled data on other architectures (Conv-NN instead of Conv)DD · CIFAR-10 (10 Img/Cls)cross-architecture accuracy (%): 60.2 ± 0.4 -> 17.8 ± 2.7Hurt
evaluate DC-distilled data on Conv-NN instead of ConvDC · CIFAR-10 (10 Img/Cls)cross-architecture accuracy (%): 44.9 ± 0.5 -> 31.9 ± 0.6Hurt
increase images per class from 1 to 50 for MTTMTT · CIFAR-100test accuracy (%): 24.3 ± 0.3 -> 47.7 ± 0.2Improved
use FRePo instead of MTT at 50 Img/Clsdataset distillation · CIFAR-100 (50 Img/Cls)test accuracy (%): 47.7 ± 0.2 -> 44.3 ± 0.2Hurt

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