DIVA: Domain Invariant Variational Autoencoders

Abstract

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.

arXiv commentCode available at https://github.com/AMLab-Amsterdam/DIVA

References in corpus (22)

• Adam: A Method for Stochastic Optimization
• Auto-Encoding Variational Bayes
• Stochastic Backpropagation and Approximate Inference in Deep Generative Models
• Weight Normalization: A Simple Reparameterization to Accelerate Training of Deep Neural Networks
• Challenging Common Assumptions in the Unsupervised Learning of Disentangled Representations
• Domain Generalization via Invariant Feature Representation
• Generalizing Across Domains via Cross-Gradient Training
• Towards a Definition of Disentangled Representations
• Identity Mappings in Deep Residual Networks
• The Variational Fair Autoencoder
• Why do deep convolutional networks generalize so poorly to small image transformations?
• On Learning Invariant Representation for Domain Adaptation
• Learning Disentangled Representations with Semi-Supervised Deep Generative Models
• Multi-Level Variational Autoencoder: Learning Disentangled Representations from Grouped Observations
• BIVA: A Very Deep Hierarchy of Latent Variables for Generative Modeling
• Learning to Generalize: Meta-Learning for Domain Generalization
• Invariant Representations without Adversarial Training
• Diagnosing and Enhancing VAE Models
• Excessive Invariance Causes Adversarial Vulnerability
• i-RevNet: Deep Invertible Networks
• Domain Generalization via Conditional Invariant Representation
• Semi-Conditional Normalizing Flows for Semi-Supervised Learning

Cited by in corpus (24)

• In Search of Lost Domain Generalization
• Domain Generalization via Optimal Transport with Metric Similarity Learning
• Lifelong Teacher-Student Network Learning
• Learning Causal Semantic Representation for Out-of-Distribution Prediction
• A Causal View on Robustness of Neural Networks
• Continuous Domain Adaptation with Variational Domain-Agnostic Feature Replay
• Selecting Data Augmentation for Simulating Interventions
• Semi-Conditional Normalizing Flows for Semi-Supervised Learning
• Semi-Supervised Generative Modeling for Controllable Speech Synthesis
• Domain Invariant Representation Learning with Domain Density Transformations
• Guided Variational Autoencoder for Disentanglement Learning
• Latent Causal Invariant Model
• Disentangling and Learning Robust Representations with Natural Clustering
• Respecting Domain Relations: Hypothesis Invariance for Domain Generalization
• Causal Curiosity: RL Agents Discovering Self-supervised Experiments for Causal Representation Learning
• Aggregating From Multiple Target-Shifted Sources
• Capturing Label Characteristics in VAEs
• Orientation-Disentangled Unsupervised Representation Learning for Computational Pathology
• Model-Based Domain Generalization
• A Bit More Bayesian: Domain-Invariant Learning with Uncertainty
• On learning latent dynamics of the AUG plasma state
• Hierarchical Variational Auto-Encoding for Unsupervised Domain Generalization
• On the Limitations of Multimodal VAEs
• Discriminative Domain-Invariant Adversarial Network for Deep Domain Generalization