DeepAI
Log In Sign Up

Batch Normalization Embeddings for Deep Domain Generalization

11/25/2020
by   Mattia Segù, et al.
62

Domain generalization aims at training machine learning models to perform robustly across different and unseen domains. Several recent methods use multiple datasets to train models to extract domain-invariant features, hoping to generalize to unseen domains. Instead, first we explicitly train domain-dependant representations by using ad-hoc batch normalization layers to collect independent domain's statistics. Then, we propose to use these statistics to map domains in a shared latent space, where membership to a domain can be measured by means of a distance function. At test time, we project samples from an unknown domain into the same space and infer properties of their domain as a linear combination of the known ones. We apply the same mapping strategy at training and test time, learning both a latent representation and a powerful but lightweight ensemble model. We show a significant increase in classification accuracy over current state-of-the-art techniques on popular domain generalization benchmarks: PACS, Office-31 and Office-Caltech.

READ FULL TEXT

page 1

page 2

page 3

page 4

05/07/2021

Adaptive Domain-Specific Normalization for Generalizable Person Re-Identification

Although existing person re-identification (Re-ID) methods have shown im...
10/26/2022

Trade-off between reconstruction loss and feature alignment for domain generalization

Domain generalization (DG) is a branch of transfer learning that aims to...
10/06/2021

Test-time Batch Statistics Calibration for Covariate Shift

Deep neural networks have a clear degradation when applying to the unsee...
11/18/2019

Domain Generalization Using a Mixture of Multiple Latent Domains

When domains, which represent underlying data distributions, vary during...
02/20/2021

On Calibration and Out-of-domain Generalization

Out-of-domain (OOD) generalization is a significant challenge for machin...
06/20/2020

Representation via Representations: Domain Generalization via Adversarially Learned Invariant Representations

We investigate the power of censoring techniques, first developed for le...
01/21/2021

Out-of-Distribution Generalization Analysis via Influence Function

The mismatch between training and target data is one major challenge for...