Incremental Learning on ImageNet - 10 steps

10개 결과 · ⬇ CSV · JSON

Average Incremental Accuracy

Rank	Model	Average Incremental Accuracy	Final Accuracy	Average Incremental Accuracy Top-5	Final Accuracy Top-5	# M Params	Extra Training Data	Paper	Code	Year
1	kNN-CLIP	85.5	–	–	–	–	✓	Revisiting a kNN-based Image Classification System with High-capacity Storage		2022
2	DyTox	71.29	63.34	88.59	84.49	11.36		DyTox: Transformers for Continual Learning with DYnamic TOken eXpansion	arthurdouillard/dytox	2021
3	DER w/o Pruning	68.84	60.16	88.17	82.86	116.89		DER: Dynamically Expandable Representation for Class Incremental Learning	g-u-n/pycil · Rhyssiyan/DER-ClassIL.pytorch	2021
4	FOSTER	68.34	–	–	–	–		FOSTER: Feature Boosting and Compression for Class-Incremental Learning	g-u-n/pycil · G-U-N/ECCV22-FOSTER	2022
5	RMM (ResNet-18)	67.45	–	–	–	–		RMM: Reinforced Memory Management for Class-Incremental Learning	g-u-n/pycil · aimagelab/mammoth · yaoyaoliu/rmm · +1	2023
6	DER	66.73	58.62	87.08	81.89	–		DER: Dynamically Expandable Representation for Class Incremental Learning	g-u-n/pycil · Rhyssiyan/DER-ClassIL.pytorch	2021
7	WA	65.67	55.60	86.60	81.10	11.68		Maintaining Discrimination and Fairness in Class Incremental Learning	g-u-n/pycil · wannaa/c-flat	2019
8	iCaRL	38.40	22.70	63.70	44.00	11.68		iCaRL: Incremental Classifier and Representation Learning	ContinualAI/avalanche · g-u-n/pycil · aimagelab/mammoth · +7	2016
9	BiC	–	–	84.00	73.20	11.68		Large Scale Incremental Learning	ContinualAI/avalanche · g-u-n/pycil · aimagelab/mammoth · +2	2019
10	E2E	–	–	72.09	52.29	11.68		End-to-End Incremental Learning	mmasana/FACIL · kibok90/iccv2019-inc · fmcp/EndToEndIncrementalLearning · +3	2018