AI Paper Atlas

Research map / Label-efficient and robust learning / Zero-/few-shot and meta-learning

Learngene: From Open-World to Your Learning Task

Qiu-Feng Wang, Xin Geng, Shu-Xia Lin, Shi-Yu Xia, Lei Qi, Ning Xu

AAAI 2022 · Label-efficient and robust learning › Zero-/few-shot and meta-learning · approach: Normalization and unsupervised meta-learning · 12 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
use learngene initialization and adaptation instead of DeepEMDLearngene individual model · CIFAR100 open-world classesaccuracy: 39.85±0.41 -> 45.88±2.99Improved
use learngene initialization and adaptation instead of Fine-tuningLearngene individual model · ImageNet100 open-world classesaccuracy: 40.72±1.36 -> 42.40±2.35Improved
use learngene initialization and adaptation instead of DeepEMDLearngene individual model · CIFAR100 5-way 10-shot novel classesaccuracy: 58.03±1.85 -> 58.86±3.12Improved
use learngene initialization and adaptation instead of DeepEMDLearngene individual model · CIFAR100 5-way 20-shot novel classesaccuracy: 63.69±2.05 -> 64.24±1.89Improved
use learngene initialization and adaptation instead of MatchingNetLearngene individual model · ImageNet100 5-way 10-shot novel classesaccuracy: 51.82±2.58 -> 52.96±4.84Improved
use learngene initialization and adaptation instead of Fine-tuningLearngene individual model · ImageNet100 5-way 20-shot novel classesaccuracy: 60.71±2.32 -> 60.86±4.48Improved
increase the retain-loss weight λ from 1000 to 5000 (ablation)Learngene individual model · CIFAR100 5-way 10-shot novel classesaccuracy: 58.86±3.12 -> 58.21±2.58Hurt
increase the retain-loss weight λ from 1000 to 5000 (ablation)Learngene individual model · CIFAR100 5-way 20-shot novel classesaccuracy: 64.24±1.89 -> 63.74±2.46Hurt
increase the retain-loss weight λ from 1000 to 5000 (ablation)Learngene individual model · ImageNet100 5-way 10-shot novel classesaccuracy: 52.96±4.84 -> 52.53±4.74Hurt
increase the retain-loss weight λ from 1000 to 5000 (ablation)Learngene individual model · ImageNet100 5-way 20-shot novel classesaccuracy: 60.86±4.48 -> 59.76±5.41Hurt
select the top network layers rather than front or middle layers as learngene (ablation)Learngene individual model · network-position comparisonaccuracy: 2% accuracy higherImproved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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