Universal Domain Adaptation through Self Supervision

02/19/2020
by   Kuniaki Saito, et al.
15

Unsupervised domain adaptation methods traditionally assume that all source categories are present in the target domain. In practice, little may be known about the category overlap between the two domains. While some methods address target settings with either partial or open-set categories, they assume that the particular setting is known a priori. We propose a more universally applicable domain adaptation approach that can handle arbitrary category shift, called Domain Adaptative Neighborhood Clustering via Entropy optimization (DANCE). DANCE combines two novel ideas: First, as we cannot fully rely on source categories to learn features discriminative for the target, we propose a novel neighborhood clustering technique to learn the structure of the target domain in a self-supervised way. Second, we use entropy-based feature alignment and rejection to align target features with the source, or reject them as unknown categories based on their entropy. We show through extensive experiments that DANCE outperforms baselines across open-set, open-partial and partial domain adaptation settings.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
12/16/2021

UMAD: Universal Model Adaptation under Domain and Category Shift

Learning to reject unknown samples (not present in the source classes) i...
research
06/11/2020

Exploring Category-Agnostic Clusters for Open-Set Domain Adaptation

Unsupervised domain adaptation has received significant attention in rec...
research
06/22/2021

Universal Domain Adaptation in Ordinal Regression

We address the problem of universal domain adaptation (UDA) in ordinal r...
research
05/06/2021

Towards Novel Target Discovery Through Open-Set Domain Adaptation

Open-set domain adaptation (OSDA) considers that the target domain conta...
research
03/23/2022

Generic network for domain adaptation based on self-supervised learning and deep clustering

Domain adaptation methods train a model to find similar feature represen...
research
03/07/2022

Open Set Domain Adaptation By Novel Class Discovery

In Open Set Domain Adaptation (OSDA), large amounts of target samples ar...
research
04/24/2023

Universal Domain Adaptation via Compressive Attention Matching

Universal domain adaptation (UniDA) aims to transfer knowledge from the ...

Please sign up or login with your details

Forgot password? Click here to reset