TTTFlow: Unsupervised Test-Time Training with Normalizing Flow
- URL: http://arxiv.org/abs/2210.11389v1
- Date: Thu, 20 Oct 2022 16:32:06 GMT
- Title: TTTFlow: Unsupervised Test-Time Training with Normalizing Flow
- Authors: David Osowiechi, Gustavo A. Vargas Hakim, Mehrdad Noori, Milad
Cheraghalikhani, Ismail Ben Ayed, Christian Desrosiers
- Abstract summary: 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.
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.
- Score: 18.121961548745112
- License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
- 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.
Related papers
- Meta-TTT: A Meta-learning Minimax Framework For Test-Time Training [5.9631503543049895]
Test-time domain adaptation is a challenging task that aims to adapt a pre-trained model to limited, unlabeled target data during inference.
This paper introduces a meta-learning minimax framework for test-time training on batch normalization layers.
arXiv Detail & Related papers (2024-10-02T16:16:05Z) - Enhancing Test Time Adaptation with Few-shot Guidance [35.13317598777832]
Deep neural networks often encounter significant performance drops while facing with domain shifts between training (source) and test (target) data.
Test Time Adaptation (TTA) methods have been proposed to adapt pre-trained source model to handle out-of-distribution streaming target data.
We develop Few-Shot Test Time Adaptation (FS-TTA), a novel and practical setting that utilizes a few-shot support set on top of TTA.
arXiv Detail & Related papers (2024-09-02T15:50:48Z) - What, How, and When Should Object Detectors Update in Continually
Changing Test Domains? [34.13756022890991]
Test-time adaptation algorithms have been proposed to adapt the model online while inferring test data.
We propose a novel online adaption approach for object detection in continually changing test domains.
Our approach surpasses baselines on widely used benchmarks, achieving improvements of up to 4.9%p and 7.9%p in mAP.
arXiv Detail & Related papers (2023-12-12T07:13:08Z) - Open Domain Generalization with a Single Network by Regularization
Exploiting Pre-trained Features [37.518025833882334]
Open Domain Generalization (ODG) is a challenging task as it deals with distribution shifts and category shifts.
Previous work has used multiple source-specific networks, which involve a high cost.
This paper proposes a method that can handle ODG using only a single network.
arXiv Detail & Related papers (2023-12-08T16:22:10Z) - Activate and Reject: Towards Safe Domain Generalization under Category
Shift [71.95548187205736]
We study a practical problem of Domain Generalization under Category Shift (DGCS)
It aims to simultaneously detect unknown-class samples and classify known-class samples in the target domains.
Compared to prior DG works, we face two new challenges: 1) how to learn the concept of unknown'' during training with only source known-class samples, and 2) how to adapt the source-trained model to unseen environments.
arXiv Detail & Related papers (2023-10-07T07:53:12Z) - Improved Test-Time Adaptation for Domain Generalization [48.239665441875374]
Test-time training (TTT) adapts the learned model with test data.
This work addresses two main factors: selecting an appropriate auxiliary TTT task for updating and identifying reliable parameters to update during the test phase.
We introduce additional adaptive parameters for the trained model, and we suggest only updating the adaptive parameters during the test phase.
arXiv Detail & Related papers (2023-04-10T10:12:38Z) - TeST: Test-time Self-Training under Distribution Shift [99.68465267994783]
Test-Time Self-Training (TeST) is a technique that takes as input a model trained on some source data and a novel data distribution at test time.
We find that models adapted using TeST significantly improve over baseline test-time adaptation algorithms.
arXiv Detail & Related papers (2022-09-23T07:47:33Z) - CAFA: Class-Aware Feature Alignment for Test-Time Adaptation [50.26963784271912]
Test-time adaptation (TTA) aims to address this challenge by adapting a model to unlabeled data at test time.
We propose a simple yet effective feature alignment loss, termed as Class-Aware Feature Alignment (CAFA), which simultaneously encourages a model to learn target representations in a class-discriminative manner.
arXiv Detail & Related papers (2022-06-01T03:02:07Z) - MT3: Meta Test-Time Training for Self-Supervised Test-Time Adaption [69.76837484008033]
An unresolved problem in Deep Learning is the ability of neural networks to cope with domain shifts during test-time.
We combine meta-learning, self-supervision and test-time training to learn to adapt to unseen test distributions.
Our approach significantly improves the state-of-the-art results on the CIFAR-10-Corrupted image classification benchmark.
arXiv Detail & Related papers (2021-03-30T09:33:38Z) - Adaptive Risk Minimization: Learning to Adapt to Domain Shift [109.87561509436016]
A fundamental assumption of most machine learning algorithms is that the training and test data are drawn from the same underlying distribution.
In this work, we consider the problem setting of domain generalization, where the training data are structured into domains and there may be multiple test time shifts.
We introduce the framework of adaptive risk minimization (ARM), in which models are directly optimized for effective adaptation to shift by learning to adapt on the training domains.
arXiv Detail & Related papers (2020-07-06T17:59:30Z)
This list is automatically generated from the titles and abstracts of the papers in this site.
This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.