Learning to Adapt to Online Streams with Distribution Shifts
- URL: http://arxiv.org/abs/2303.01630v1
- Date: Thu, 2 Mar 2023 23:36:10 GMT
- Title: Learning to Adapt to Online Streams with Distribution Shifts
- Authors: Chenyan Wu, Yimu Pan, Yandong Li, James Z. Wang
- Abstract summary: Test-time adaptation (TTA) is a technique used to reduce distribution gaps between the training and testing sets by leveraging unlabeled test data during inference.
In this work, we expand TTA to a more practical scenario, where the test data comes in the form of online streams that experience distribution shifts over time.
We propose a meta-learning approach that teaches the network to adapt to distribution-shifting online streams during meta-training. As a result, the trained model can perform continual adaptation to distribution shifts in testing, regardless of the batch size restriction.
- Score: 22.155844301575883
- License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
- Abstract: Test-time adaptation (TTA) is a technique used to reduce distribution gaps
between the training and testing sets by leveraging unlabeled test data during
inference. In this work, we expand TTA to a more practical scenario, where the
test data comes in the form of online streams that experience distribution
shifts over time. Existing approaches face two challenges: reliance on a large
test data batch from the same domain and the absence of explicitly modeling the
continual distribution evolution process. To address both challenges, we
propose a meta-learning approach that teaches the network to adapt to
distribution-shifting online streams during meta-training. As a result, the
trained model can perform continual adaptation to distribution shifts in
testing, regardless of the batch size restriction, as it has learned during
training. We conducted extensive experiments on benchmarking datasets for TTA,
incorporating a broad range of online distribution-shifting settings. Our
results showed consistent improvements over state-of-the-art methods,
indicating the effectiveness of our approach. In addition, we achieved superior
performance in the video segmentation task, highlighting the potential of our
method for real-world applications.
Related papers
- Distribution Alignment for Fully Test-Time Adaptation with Dynamic Online Data Streams [19.921480334048756]
Test-Time Adaptation (TTA) enables adaptation and inference in test data streams with domain shifts from the source.
We propose a novel Distribution Alignment loss for TTA.
We surpass existing methods in non-i.i.d. scenarios and maintain competitive performance under the ideal i.i.d. assumption.
arXiv Detail & Related papers (2024-07-16T19:33:23Z) - Any-Shift Prompting for Generalization over Distributions [66.29237565901734]
We propose any-shift prompting: a general probabilistic inference framework that considers the relationship between training and test distributions during prompt learning.
Within this framework, the test prompt exploits the distribution relationships to guide the generalization of the CLIP image-language model from training to any test distribution.
The network generates the tailored test prompt with both training and test information in a feedforward pass, avoiding extra training costs at test time.
arXiv Detail & Related papers (2024-02-15T16:53:42Z) - Consistency Regularization for Generalizable Source-free Domain
Adaptation [62.654883736925456]
Source-free domain adaptation (SFDA) aims to adapt a well-trained source model to an unlabelled target domain without accessing the source dataset.
Existing SFDA methods ONLY assess their adapted models on the target training set, neglecting the data from unseen but identically distributed testing sets.
We propose a consistency regularization framework to develop a more generalizable SFDA method.
arXiv Detail & Related papers (2023-08-03T07:45:53Z) - Conditional Online Learning for Keyword Spotting [0.0]
This work investigates a simple but effective online continual learning method that updates a keyword spotter on-device via SGD as new data becomes available.
Experiments demonstrate that, compared to a naive online learning implementation, conditional model updates based on its performance in a small hold-out set drawn from the training distribution mitigate catastrophic forgetting.
arXiv Detail & Related papers (2023-05-19T15:46:31Z) - Test-Time Adaptation with Perturbation Consistency Learning [32.58879780726279]
We propose a simple test-time adaptation method to promote the model to make stable predictions for samples with distribution shifts.
Our method can achieve higher or comparable performance with less inference time over strong PLM backbones.
arXiv Detail & Related papers (2023-04-25T12:29:22Z) - A Comprehensive Survey on Test-Time Adaptation under Distribution Shifts [143.14128737978342]
Test-time adaptation, an emerging paradigm, has the potential to adapt a pre-trained model to unlabeled data during testing, before making predictions.
Recent progress in this paradigm highlights the significant benefits of utilizing unlabeled data for training self-adapted models prior to inference.
arXiv Detail & Related papers (2023-03-27T16:32:21Z) - Addressing Distribution Shift at Test Time in Pre-trained Language
Models [3.655021726150369]
State-of-the-art pre-trained language models (PLMs) outperform other models when applied to the majority of language processing tasks.
PLMs have been found to degrade in performance under distribution shift.
We present an approach that improves the performance of PLMs at test-time under distribution shift.
arXiv Detail & Related papers (2022-12-05T16:04:54Z) - 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) - VisDA-2021 Competition Universal Domain Adaptation to Improve
Performance on Out-of-Distribution Data [64.91713686654805]
The Visual Domain Adaptation (VisDA) 2021 competition tests models' ability to adapt to novel test distributions.
We will evaluate adaptation to novel viewpoints, backgrounds, modalities and degradation in quality.
Performance will be measured using a rigorous protocol, comparing to state-of-the-art domain adaptation methods.
arXiv Detail & Related papers (2021-07-23T03:21:51Z) - Unsupervised neural adaptation model based on optimal transport for
spoken language identification [54.96267179988487]
Due to the mismatch of statistical distributions of acoustic speech between training and testing sets, the performance of spoken language identification (SLID) could be drastically degraded.
We propose an unsupervised neural adaptation model to deal with the distribution mismatch problem for SLID.
arXiv Detail & Related papers (2020-12-24T07:37:19Z)
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.