TY - GEN
T1 - Partial Domain Adaptation Using Selective Representation Learning for Class-Weight Computation
AU - Choudhuri, Sandipan
AU - Paul, Riti
AU - Sen, Arunabha
AU - Li, Baoxin
AU - Venkateswara, Hemanth
N1 - Publisher Copyright:
© 2020 IEEE.
PY - 2020/11/1
Y1 - 2020/11/1
N2 - The generalization power of deep-learning models is dependent on rich-labelled data. This supervision using large-scaled annotated information is restrictive in most realworld scenarios where data collection and their annotation involve huge cost. Various domain adaptation techniques exist in literature that bridge this distribution discrepancy. However, a majority of these models require the label sets of both the domains to be identical. To tackle a more practical and challenging scenario, we formulate the problem statement from a partial domain adaptation perspective, where the source label set is a super set of the target label set. Driven by the motivation that image styles are private to each domain, in this work, we develop a method that identifies outlier classes exclusively from image content information and train a label classifier exclusively on class-content from source images. Additionally, elimination of negative transfer of samples from classes private to the source domain is achieved by transforming the soft class-level weights into two clusters, 0 (outlier source classes) and 1 (shared classes) by maximizing the between-cluster variance between them.
AB - The generalization power of deep-learning models is dependent on rich-labelled data. This supervision using large-scaled annotated information is restrictive in most realworld scenarios where data collection and their annotation involve huge cost. Various domain adaptation techniques exist in literature that bridge this distribution discrepancy. However, a majority of these models require the label sets of both the domains to be identical. To tackle a more practical and challenging scenario, we formulate the problem statement from a partial domain adaptation perspective, where the source label set is a super set of the target label set. Driven by the motivation that image styles are private to each domain, in this work, we develop a method that identifies outlier classes exclusively from image content information and train a label classifier exclusively on class-content from source images. Additionally, elimination of negative transfer of samples from classes private to the source domain is achieved by transforming the soft class-level weights into two clusters, 0 (outlier source classes) and 1 (shared classes) by maximizing the between-cluster variance between them.
UR - http://www.scopus.com/inward/record.url?scp=85107777491&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85107777491&partnerID=8YFLogxK
U2 - 10.1109/IEEECONF51394.2020.9443420
DO - 10.1109/IEEECONF51394.2020.9443420
M3 - Conference contribution
AN - SCOPUS:85107777491
T3 - Conference Record - Asilomar Conference on Signals, Systems and Computers
SP - 289
EP - 293
BT - Conference Record of the 54th Asilomar Conference on Signals, Systems and Computers, ACSSC 2020
A2 - Matthews, Michael B.
PB - IEEE Computer Society
T2 - 54th Asilomar Conference on Signals, Systems and Computers, ACSSC 2020
Y2 - 1 November 2020 through 5 November 2020
ER -