Support and invertibility in domain-invariant representations
Learning domain-invariant representations has become a popular approach to unsupervised domain adaptation and is often justified by invoking a particular suite of theoretical results. We argue that there are two significant flaws in such arguments. First, the results in question hold only for a fixe...
Main Authors: | Johansson, Fredrik D. (Author), Sontag, David Alexander (Author) |
---|---|
Other Authors: | Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science (Contributor) |
Format: | Article |
Language: | English |
Published: |
International Machine Learning Society,
2021-04-05T14:00:53Z.
|
Subjects: | |
Online Access: | Get fulltext |
Similar Items
-
Learning Domain-Invariant Representations of Histological Images
by: Maxime W. Lafarge, et al.
Published: (2019-07-01) -
Neural representations that support invariant object recognition
by: Robbe L. T Goris, et al.
Published: (2009-02-01) -
Homological invariants of strongly invertible knots
by: Snape, Michael
Published: (2018) -
Invariance to representation of information
by: Yildiz, Muhamet
Published: (2018) -
Invariable copula in inverted cleft sentences: an example of grammaticalization?
by: Damaris Matias Silveira
Published: (2016-12-01)