DIVA: Domain Invariant Variational Autoencoders

05/24/2019
by   Maximilian Ilse, et al.
11

We consider the problem of domain generalization, namely, how to learn representations given data from a set of domains that generalize to data from a previously unseen domain. We propose the Domain Invariant Variational Autoencoder (DIVA), a generative model that tackles this problem by learning three independent latent subspaces, one for the domain, one for the class, and one for any residual variations. We highlight that due to the generative nature of our model we can also incorporate unlabeled data from known or previously unseen domains. To the best of our knowledge this has not been done before in a domain generalization setting. This property is highly desirable in fields like medical imaging where labeled data is scarce. We experimentally evaluate our model on the rotated MNIST benchmark and a malaria cell images dataset where we show that (i) the learned subspaces are indeed complementary to each other, (ii) we improve upon recent works on this task and (iii) incorporating unlabelled data can boost the performance even further.

READ FULL TEXT

page 7

page 13

research
01/23/2021

Hierarchical Domain Invariant Variational Auto-Encoding with weak domain supervision

We address the task of domain generalization, where the goal is to train...
research
02/09/2021

Domain Invariant Representation Learning with Domain Density Transformations

Domain generalization refers to the problem where we aim to train a mode...
research
02/25/2021

Model-Based Domain Generalization

We consider the problem of domain generalization, in which a predictor i...
research
07/15/2020

Learning to Learn with Variational Information Bottleneck for Domain Generalization

Domain generalization models learn to generalize to previously unseen do...
research
09/20/2023

Generalizing Across Domains in Diabetic Retinopathy via Variational Autoencoders

Domain generalization for Diabetic Retinopathy (DR) classification allow...
research
08/31/2015

Domain Generalization for Object Recognition with Multi-task Autoencoders

The problem of domain generalization is to take knowledge acquired from ...
research
01/15/2021

Harmonization and the Worst Scanner Syndrome

We show that for a wide class of harmonization/domain-invariance schemes...

Please sign up or login with your details

Forgot password? Click here to reset