Understanding Test-Time Augmentation

2024/02/10 by Masanari Kimura, Kimura, Masanari · 9 citations
Engineering · #Engineering and Test Systems #FOS: Computer and information sciences #Machine Learning (cs.LG)

paper · pdf · doi:10.48550/arxiv.2402.06892

openalex publication_date 2024/02/10 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

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.

Cited by

• AVP-Fusion: Adaptive Multi-Modal Fusion and Contrastive Learning for Two-Stage Antiviral Peptide Identification
• RoboDriveVLM: A Novel Benchmark and Baseline towards Robust Vision-Language Models for Autonomous Driving
• Reliable Robotic Task Execution in the Face of Anomalies
• AMAuT: A Flexible and Efficient Multiview Audio Transformer Framework Trained from Scratch
• prNet: Data-Driven Phase Retrieval via Stochastic Refinement
• Learning from Random Subspace Exploration: Generalized Test-Time Augmentation with Self-supervised Distillation
• Higher-Order Asymptotics of Test-Time Adaptation for Batch Normalization Statistics
• Test-Time Augmentation for Pose-invariant Face Recognition
• Test-Time Scaling for Safe Text-Guided Image Generation via Intermediate Clean Estimates

Related

• Test-Time Augmentation Meets Variational Bayes
• Improved Text Classification via Test-Time Augmentation
• Learning Loss for Test-Time Augmentation
• How Much Data Are Augmentations Worth? An Investigation into Scaling Laws, Invariance, and Implicit Regularization
• On the Adversarial Risk of Test Time Adaptation: An Investigation into Realistic Test-Time Data Poisoning