Conference item icon

Conference item

Turning a blind eye: Explicit removal of biases and variation from deep neural network embeddings

Abstract:

Neural networks achieve the state-of-the-art in image classification tasks. However, they can encode spurious variations or biases that may be present in the training data. For example, training an age predictor on a dataset that is not balanced for gender can lead to gender biased predicitons (e.g. wrongly predicting that males are older if only elderly males are in the training set). We present two distinct contributions:

(1) An algorithm that can remove multiple sources of v...

Expand abstract
Publication status:
Published
Peer review status:
Reviewed (other)
Version:
Accepted Manuscript

Actions


Access Document


Files:
Publisher copy:
10.1007/978-3-030-11009-3_34

Authors


More by this author
Institution:
University of Oxford
Division:
MPLS Division
Department:
Department of Engineering Science
More by this author
Institution:
University of Oxford
Division:
MPLS Division
Department:
Department of Engineering Science
Oxford college:
Brasenose College
ORCID:
0000-0002-8945-8573
More by this author
Institution:
University of Oxford
Division:
Medical Sciences Division
Department:
Women's and Reproductive Health
ORCID:
0000-0002-2887-2068
Publisher:
Springer Publisher's website
Volume:
11129
Pages:
556-572
Series:
Lecture Notes in Computer Science
Publication date:
2019-01-23
Acceptance date:
2018-08-29
DOI:
EISSN:
1611-3349
ISSN:
0302-9743
Pubs id:
pubs:921882
URN:
uri:ef627761-c2f7-47aa-89d1-f51d99e07285
UUID:
uuid:ef627761-c2f7-47aa-89d1-f51d99e07285
Local pid:
pubs:921882
ISBN:
9783030110086

Terms of use


Metrics



If you are the owner of this record, you can report an update to it here: Report update to this record

TO TOP