HyperAI초신경

홈 뉴스 연구 논문 튜토리얼 데이터셋 백과사전 SOTA LLM 모델 GPU 랭킹 컨퍼런스

한국어

HyperAI초신경

Continual Learning On Visual Domain Decathlon

평가 지표

decathlon discipline (Score)

평가 결과

이 벤치마크에서 각 모델의 성능 결과

모델 이름	decathlon discipline (Score)	Paper Title	Repository
Res. adapt. finetune all	2643	Learning multiple visual domains with residual adapters
Piggyback	2838	Piggyback: Adapting a Single Network to Multiple Tasks by Learning to Mask Weights
Res. adapt.	2118	Learning multiple visual domains with residual adapters
DAN	2851	Incremental Learning Through Deep Adaptation	-
Res. adapt. decay	2621	Learning multiple visual domains with residual adapters
NetTailor	3744	NetTailor: Tuning the Architecture, Not Just the Weights
Res. adapt. dom-pred	2503	Learning multiple visual domains with residual adapters
Series Res. adapt.	3159	Efficient parametrization of multi-domain deep neural networks
LwF	2515	Learning without Forgetting
BN adapt.	1363	Universal representations:The missing link between faces, text, planktons, and cat breeds	-
Res. adapt. (large)	3131	Learning multiple visual domains with residual adapters
Depthwise Sharing	3234	Depthwise Convolution is All You Need for Learning Multiple Visual Domains
Depthwise Soft Sharing	3507	Depthwise Convolution is All You Need for Learning Multiple Visual Domains
Parallel Res. adapt.	3412	Efficient parametrization of multi-domain deep neural networks

0 of 14 row(s) selected.