HyperAI초신경

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

한국어

HyperAI초신경

Continual Learning On Asc 19 Tasks

평가 지표

F1 - macro

평가 결과

이 벤치마크에서 각 모델의 성능 결과

모델 이름	F1 - macro	Paper Title	Repository
OWM	0.7931	Continual Learning of Context-dependent Processing in Neural Networks
CTR	0.8362	Achieving Forgetting Prevention and Knowledge Transfer in Continual Learning
CAT	0.6864	Continual Learning of a Mixed Sequence of Similar and Dissimilar Tasks
B-CL	0.8140	Adapting BERT for Continual Learning of a Sequence of Aspect Sentiment Classification Tasks
KAN	0.7738	Continual Learning with Knowledge Transfer for Sentiment Classification
A-GEM	0.7844	Efficient Lifelong Learning with A-GEM
L2	0.5243	Overcoming catastrophic forgetting in neural networks
UCL	0.7599	Uncertainty-based Continual Learning with Adaptive Regularization
DER++	0.7508	Dark Experience for General Continual Learning: a Strong, Simple Baseline
EWC	0.7452	Overcoming catastrophic forgetting in neural networks
HAT	0.7816	Overcoming catastrophic forgetting with hard attention to the task
Naive Continual Learning (NCL)	0.7664	Achieving Forgetting Prevention and Knowledge Transfer in Continual Learning
LAMOL	0.8059	LAMOL: LAnguage MOdeling for Lifelong Language Learning
Multi-task Learning (MTL; Upper Bound)	0.8811	Achieving Forgetting Prevention and Knowledge Transfer in Continual Learning
Independent Learning (ONE)	0.7807	Achieving Forgetting Prevention and Knowledge Transfer in Continual Learning

0 of 15 row(s) selected.