Sökning: onr:"swepub:oai:DiVA.org:kth-251684" >
Toward principled r...
Toward principled regularization of deep networks : From weight decay to feature contraction
-
- Maki, Atsuto (författare)
- KTH,Robotik, perception och lärande, RPL
-
(creator_code:org_t)
- American Association for the Advancement of Science (AAAS), 2019
- 2019
- Engelska.
-
Ingår i: Science Robotics. - : American Association for the Advancement of Science (AAAS). - 2470-9476. ; 4:30
- Relaterad länk:
-
https://robotics.sci...
-
visa fler...
-
https://urn.kb.se/re...
-
https://doi.org/10.1...
-
visa färre...
Abstract
Ämnesord
Stäng
- Training deep artificial neural networks for classification problems may benefit from exploiting intrinsic class similarities by way of network regularization that compensates for a drawback in the commonly used target error.
Ämnesord
- TEKNIK OCH TEKNOLOGIER -- Elektroteknik och elektronik (hsv//swe)
- ENGINEERING AND TECHNOLOGY -- Electrical Engineering, Electronic Engineering, Information Engineering (hsv//eng)
Nyckelord
- Computer Science
- Datalogi
Publikations- och innehållstyp
- ref (ämneskategori)
- art (ämneskategori)
Hitta via bibliotek
Till lärosätets databas