English

Understanding Test-Time Augmentation

Machine Learning 2024-02-13 v1
Authors: Masanari Kimura
View on arXiv ↗ PDF ↗

Abstract

Test-Time Augmentation (TTA) is a very powerful heuristic that takes advantage of data augmentation during testing to produce averaged output. Despite the experimental effectiveness of TTA, there is insufficient discussion of its theoretical aspects. In this paper, we aim to give theoretical guarantees for TTA and clarify its behavior.

Keywords

test-time adaptation

Cite

@article{arxiv.2402.06892,
  title  = {Understanding Test-Time Augmentation},
  author = {Masanari Kimura},
  journal= {arXiv preprint arXiv:2402.06892},
  year   = {2024}
}

Related papers

View all related →
Machine Learning · Statistics
Test-Time Augmentation Meets Variational Bayes
Masanari Kimura, Howard Bondell
2024-09-20
Machine Learning · Computer Science
Improved Text Classification via Test-Time Augmentation
Helen Lu, Divya Shanmugam, Harini Suresh, John Guttag
2022-06-29
Machine Learning · Computer Science
Test-time augmentation improves efficiency in conformal prediction
Divya Shanmugam, Helen Lu, Swami Sankaranarayanan, John Guttag
2025-05-30
Computation and Language · Computer Science
Test-time Augmentation for Factual Probing
Go Kamoda, Benjamin Heinzerling, Keisuke Sakaguchi, Kentaro Inui
2023-10-27
Computer Vision and Pattern Recognition · Computer Science
I Can't Believe TTA Is Not Better: When Test-Time Augmentation Hurts Medical Image Classification
Daniel Nobrega Medeiros
2026-04-14
Computer Vision and Pattern Recognition · Computer Science
Feature Augmentation based Test-Time Adaptation
Younggeol Cho, Youngrae Kim, Junho Yoon, Seunghoon Hong +1
2024-10-21
Information Retrieval · Computer Science
Data Augmentation as Free Lunch: Exploring the Test-Time Augmentation for Sequential Recommendation
Yizhou Dang, Yuting Liu, Enneng Yang, Minhan Huang +3
2025-05-01
Machine Learning · Computer Science
Boosting Anomaly Detection Using Unsupervised Diverse Test-Time Augmentation
Seffi Cohen, Niv Goldshlager, Lior Rokach, Bracha Shapira
2025-02-10
Machine Learning · Computer Science
Active Test-Time Adaptation: Theoretical Analyses and An Algorithm
Shurui Gui, Xiner Li, Shuiwang Ji
2024-04-09
Machine Learning · Computer Science
Test-Time Augmentation for Traveling Salesperson Problem
Ryo Ishiyama, Takahiro Shirakawa, Seiichi Uchida, Shinnosuke Matsuo
2024-05-09
Image and Video Processing · Electrical Eng. & Systems
Intelligent Multi-View Test Time Augmentation
Efe Ozturk, Mohit Prabhushankar, Ghassan AlRegib
2024-06-14
Information Retrieval · Computer Science
Beyond One-Size-Fits-All: Adaptive Test-Time Augmentation for Sequential Recommendation
Xibo Li, Liang Zhang
2026-04-20
Computer Vision and Pattern Recognition · Computer Science
Improved Test-Time Adaptation for Domain Generalization
Liang Chen, Yong Zhang, Yibing Song, Ying Shan +1
2023-04-18
Computer Vision and Pattern Recognition · Computer Science
Better Aggregation in Test-Time Augmentation
Divya Shanmugam, Davis Blalock, Guha Balakrishnan, John Guttag
2021-10-13
Computer Vision and Pattern Recognition · Computer Science
Learning from Random Subspace Exploration: Generalized Test-Time Augmentation with Self-supervised Distillation
Andrei Jelea, Ahmed Nabil Belbachir, Marius Leordeanu
2026-01-01
Machine Learning · Computer Science
TEA: Test-time Energy Adaptation
Yige Yuan, Bingbing Xu, Liang Hou, Fei Sun +2
2024-02-28
Computational Engineering, Finance, and Science · Computer Science
Test-time data augmentation: improving predictions of recurrent neural network models of composites
Petter Uvdal, Mohsen Mirkhalaf
2024-09-05
Machine Learning · Computer Science
A Comprehensive Survey on Test-Time Adaptation under Distribution Shifts
Jian Liang, Ran He, Tieniu Tan
2024-12-13
Machine Learning · Computer Science
Bag of Tricks for Fully Test-Time Adaptation
Saypraseuth Mounsaveng, Florent Chiaroni, Malik Boudiaf, Marco Pedersoli +1
2023-11-13
Machine Learning · Computer Science
Improving Black-box Robustness with In-Context Rewriting
Kyle O'Brien, Nathan Ng, Isha Puri, Jorge Mendez +4
2024-08-06
Computer Vision and Pattern Recognition · Computer Science
BayTTA: Uncertainty-aware medical image classification with optimized test-time augmentation using Bayesian model averaging
Zeinab Sherkatghanad, Moloud Abdar, Mohammadreza Bakhtyari, Pawel Plawiak +1
2024-08-28
Machine Learning · Computer Science
Learning to Adapt to Online Streams with Distribution Shifts
Chenyan Wu, Yimu Pan, Yandong Li, James Z. Wang
2023-03-06
Information Theory · Computer Science
A Bit of Information Theory, and the Data Augmentation Algorithm Converges
Yaming Yu
2009-09-12
Computer Vision and Pattern Recognition · Computer Science
Backpropagation-Free Test-Time Adaptation via Probabilistic Gaussian Alignment
Youjia Zhang, Youngeun Kim, Young-Geun Choi, Hongyeob Kim +2
2026-05-04
Computer Vision and Pattern Recognition · Computer Science
Test-time Adaptation vs. Training-time Generalization: A Case Study in Human Instance Segmentation using Keypoints Estimation
Kambiz Azarian, Debasmit Das, Hyojin Park, Fatih Porikli
2022-12-14