onr:"swepub:oai:DiVA.org:liu-198763"
Search: onr:"swepub:oai:DiVA.org:liu-198763" > Enhancing Represent...
- 1 of 1
- Previous record
- Next record
- To hitlist
Enhancing Representation Learning with Deep Classifiers in Presence of Shortcut
-
- Ahmadian, Amirhossein, 1992- (author)
- Linköpings universitet,Statistik och maskininlärning,Tekniska fakulteten
-
- Lindsten, Fredrik, 1984- (author)
- Linköpings universitet,Reglerteknik,Statistik och maskininlärning,Tekniska fakulteten
- (creator_code:org_t)
- 2023
- 2023
- English.
- In: Proceedings of IEEE ICASSP 2023.
- Conference paper (peer-reviewed)
Abstract
Subject headings
Close
- A deep neural classifier trained on an upstream task can be leveraged to boost the performance of another classifier in a related downstream task through the representations learned in hidden layers. However, presence of shortcuts (easy-to-learn features) in the upstream task can considerably impair the versatility of intermediate representations and, in turn, the downstream performance. In this paper, we propose a method to improve the representations learned by deep neural image classifiers in spite of a shortcut in upstream data. In our method, the upstream classification objective is augmented with a type of adversarial training where an auxiliary network, so called lens, fools the classifier by exploiting the shortcut in reconstructing images. Empirical comparisons in self-supervised and transfer learning problems with three shortcut-biased datasets suggest the advantages of our method in terms of downstream performance and/or training time.
Subject headings
- NATURVETENSKAP -- Data- och informationsvetenskap -- Datavetenskap (hsv//swe)
- NATURAL SCIENCES -- Computer and Information Sciences -- Computer Sciences (hsv//eng)
- NATURVETENSKAP -- Data- och informationsvetenskap -- Datorseende och robotik (hsv//swe)
- NATURAL SCIENCES -- Computer and Information Sciences -- Computer Vision and Robotics (hsv//eng)
Keyword
- Deep Representation Learning
- Shortcut Learning
- Transfer Learning
- Adversarial Methods
- Computer Vision
Publication and Content Type
- ref (subject category)
- kon (subject category)
To the university's database
- 1 of 1
- Previous record
- Next record
- To hitlist
Find more in SwePub
- By the author/editor
- Ahmadian, Amirho ...
- Lindsten, Fredri ...
- About the subject
-
- NATURAL SCIENCES
- NATURAL SCIENCES
- and Computer and Inf ...
- and Computer Science ...
-
- NATURAL SCIENCES
- NATURAL SCIENCES
- and Computer and Inf ...
- and Computer Vision ...
- Articles in the publication
- Proceedings of I ...
- By the university
- Linköping University
Search outside SwePub
- Extend your search to:
- Google Book Search
- Google Scholar
Kungliga biblioteket hanterar dina personuppgifter i enlighet med EU:s dataskyddsförordning (2018), GDPR. Läs mer om hur det funkar här.
Så här hanterar KB dina uppgifter vid användning av denna tjänst.
- Copyright © LIBRIS - National Library Systems
- LIBRIS.kb.se
