Task-Incremental Learning is a continual learning paradigm in which a model learns to perform a sequence of distinct tasks incrementally, with task identity provided during both training and inference, thereby enabling the model to leverage task-specific information to mitigate catastrophic forgetting and adapt to new tasks over time.
213
Publications
20.2K
Citations
657
Authors
268
Institutions
Publications per year
2017–2026
159
Parents
Publications
213
A Neostriatal Habit Learning System in Humans
Barbara J. Knowlton, Jennifer A. Mangels, Larry R. Squire · Science · 1996 · 1.8K citations
1–5 of 213
Authors
657
- BL
9
h-index
- MM
8
h-index
- TT
7
h-index
- JV
7
h-index
- ZK
6
h-index
1–5 of 657
Institutions
268
University of Chinese Academy of Sciences
Beijing, China
9 h-index
25 publications
763 citations
Meta (Israel)
Tel Aviv, Israel
8 h-index
17 publications
857 citations
Chinese Academy of Sciences
Beijing, China
6 h-index
23 publications
889 citations
University of Illinois Chicago
Chicago, United States
5 h-index
11 publications
372 citations
Massachusetts Institute of Technology
Cambridge, United States
5 h-index
7 publications
257 citations
Peking University
Beijing, China
5 h-index
15 publications
426 citations
Amazon (Germany)
Berlin, Germany
4 h-index
10 publications
343 citations
KU Leuven
Leuven, Belgium
4 h-index
9 publications
2K citations
1–8 of 268
Venues
- arXiv (Cornell University)
Repository · 47 publications · 2.7K citations
23
h-index
- Proceedings of the AAAI Conference on Artificial Intelligence
Conference · 11 publications · 955 citations
11
h-index
- IEEE Transactions on Neural Networks and Learning Systems
7 publications · 267 citations
7
h-index
- IEEE Transactions on Pattern Analysis and Machine Intelligence
TPAMI · 7 publications · 6.2K citations
7
h-index
- 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
Conference · 5 publications · 492 citations
5
h-index
1–5