AI Paper Atlas

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

Few-Shot Dataset Distillation via Translative Pre-Training

Songhua Liu, Xinchao Wang

ICCV 2023 · Label-efficient and robust learning › Knowledge distillation · approach: Self and data-free distillation · 8 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
add translator pre-training on ImageNet before adaptation (ablation)Translator adaptation · CIFAR10 IPC 10accuracy (%): 53.8 -> 59.2Improved
add target-dataset adaptation to pre-trained translator (ablation)Pre-trained translator · CIFAR10 IPC 10accuracy (%): 54.7 -> 59.2Improved
use pre-trained translator instead of a single network without translation (ablation)1 Net distillation · CIFAR10 IPC 10accuracy (%): 50.5 -> 54.7Improved
replace translative model with bi-level algorithm (ablation)Distillation space learning · CIFAR10 IPC 10accuracy (%): 59.2 -> 52.1Hurt
initialize adaptation from ImageNet pre-trained translator rather than a reconstruction auto-encoder (ablation)Translator adaptation · CIFAR100 IPC 10accuracy (%): 27.0 -> 35.1Improved
few-shot method (one network + translator) versus many networks (ablation)Ours vs Baseline · CIFAR10 IPC 10accuracy (%): 57.9 -> 59.2Improved
use down-sampled parameterization (Ours-DS)Ours · CIFAR10 IPC 1accuracy (%): 44.0 -> 51.4Improved
few-shot translative method versus FRePoOurs (default parameterization) · CIFAR10/CIFAR100accuracy (%): about 3 to 7 points behind FRePo, with 30 to 120x accelerationHurt
adaptation on small IPC generalizes to unseen IPCAdapted translator · CIFAR10accuracy (%)Improved

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