TTTFlow: Unsupervised Test-Time Training with Normalizing Flow

TL;DR

TTTFlow introduces an unsupervised approach using Normalizing Flows for test-time adaptation in image classification, effectively detecting and adjusting to domain shifts without task-specific proxy tasks.

Contribution

The paper presents TTTFlow, a novel unsupervised test-time training method employing Normalizing Flows to detect and adapt to domain shifts without requiring proxy tasks.

Findings

Significantly improves accuracy over previous methods.

Effectively detects domain shifts using Normalizing Flows.

Does not require task-specific proxy tasks.

Abstract

A major problem of deep neural networks for image classification is their vulnerability to domain changes at test-time. Recent methods have proposed to address this problem with test-time training (TTT), where a two-branch model is trained to learn a main classification task and also a self-supervised task used to perform test-time adaptation. However, these techniques require defining a proxy task specific to the target application. To tackle this limitation, we propose TTTFlow: a Y-shaped architecture using an unsupervised head based on Normalizing Flows to learn the normal distribution of latent features and detect domain shifts in test examples. At inference, keeping the unsupervised head fixed, we adapt the model to domain-shifted examples by maximizing the log likelihood of the Normalizing Flow. Our results show that our method can significantly improve the accuracy with respect to previous works.