Selective Sharing
Selective Sharing is a multi-task learning strategy that belongs to the class of parameter-sharing methods, distinguished by its deliberate restriction of parameter reuse to specific subsets of tasks based on task-relatedness, thereby balancing parameter efficiency and task-specific adaptation.
33
Publications
2.4K
Citations
106
Authors
34
Institutions
Publications per year
2017–2026
21
Parents
33
Dynamic Scene Deblurring With Parameter Selective Sharing and Nested Skip Connections
Hongyun Gao, Xin Tao, Xiaoyong Shen et al. · 2019 · 380 citations
Learning Task Grouping and Overlap in Multi-task Learning
Abhishek Kumar, Hal Daumé · arXiv (Cornell University) · 2012 · 253 citations · Full text
Sanjay Kairam, Mike Brzozowski, David Huffaker et al. · 2012 · 172 citations
1–5 of 33
106
2
h-index
2
h-index
2
h-index
2
h-index
2
h-index
1–5 of 106
34

National Bureau of Economic Research
Cambridge, United States
2 h-index
4 publications
142 citations
Chinese University of Hong Kong
Hong Kong, Hong Kong
2 h-index
3 publications
818 citations
University of California San Diego
San Diego, United States
2 h-index
5 publications
146 citations
Michigan State University
East Lansing, United States
2 h-index
5 publications
539 citations
University of Massachusetts Dartmouth
North Dartmouth, United States
1 h-index
2 publications
30 citations
Northwestern University
Evanston, United States
1 h-index
2 publications
102 citations
University of Southern California
Los Angeles, United States
1 h-index
1 publications
479 citations
Google (United States)
Mountain View, United States
1 h-index
3 publications
516 citations
1–8 of 34
2 publications · 537 citations
2
h-index
Repository · 2 publications · 270 citations
2
h-index
2 publications · 260 citations
2
h-index
1 publications · 23 citations
1
h-index
1 publications · 14 citations
1
h-index
1–5