Sökning: WFRF:(Lindsten Fredrik 1984 ) >
Enhancing Represent...
Enhancing Representation Learning with Deep Classifiers in Presence of Shortcut
-
- Ahmadian, Amirhossein, 1992- (författare)
- Linköpings universitet,Statistik och maskininlärning,Tekniska fakulteten
-
- Lindsten, Fredrik, 1984- (författare)
- Linköpings universitet,Reglerteknik,Statistik och maskininlärning,Tekniska fakulteten
-
(creator_code:org_t)
- 2023
- 2023
- Engelska.
-
Ingår i: Proceedings of IEEE ICASSP 2023.
- Relaterad länk:
-
https://urn.kb.se/re...
-
visa fler...
-
https://doi.org/10.1...
-
visa färre...
Abstract
Ämnesord
Stäng
- 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.
Ämnesord
- 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)
Nyckelord
- Deep Representation Learning
- Shortcut Learning
- Transfer Learning
- Adversarial Methods
- Computer Vision
Publikations- och innehållstyp
- ref (ämneskategori)
- kon (ämneskategori)