AI Paper Atlas

Research map / Label-efficient and robust learning / Recognition and long-tail

TrivialAugment: Tuning-Free Yet State-of-the-Art Data Augmentation

Samuel G. Müller, Frank Hutter

ICCV 2021 · Label-efficient and robust learning › Recognition and long-tail · approach: Loss design and classifier theory · 372 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
use TA instead of brute-force RA search (91 trainings)Wide-ResNet-40-2 · CIFAR-10accuracy: 96.42 -> 96.62Improved
use Wide augmentation space instead of RA space (ablation)TA (Wide-ResNet-28-10) · CIFAR-100accuracy: 83.54 -> 84.33Improved
restrict to Full augmentation space (adds flips, blur, smooth) instead of RA space (ablation)TA (Wide-ResNet-28-10) · SVHN Coreaccuracy: 98.05 -> 97.63Hurt
use Full augmentation space instead of RA space (ablation)TA (Wide-ResNet-28-10) · CIFAR-10accuracy: 97.46 -> 97.24Hurt
remove invert from AA space (ablation)TA (Wide-ResNet-28-10) · CIFAR-10accuracy: 97.47 -> 97.55Improved
remove invert from AA space (ablation)TA (Wide-ResNet-28-10) · SVHN Coreaccuracy: 98.04 -> 97.97Hurt
sample only strength 30 instead of 0-30 (ablation)TA (RA) · CIFAR-100accuracy: 83.54 -> 82.98Hurt
sample only strength 30 instead of 0-30 (ablation)TA (RA) · SVHN Coreaccuracy: 98.05 -> 98.16Improved
sample only strength 30 instead of 0-30 (ablation)TA (RA) · CIFAR-10accuracy: 97.46 -> 97.45No clear effect
reduce strengths to {0,30} from 31 values (ablation)TA (RA) · CIFAR-100accuracy: 83.54 -> 83.46No clear effect
use TA x8 batch augmentation instead of Adv. AAWide-ResNet-28-10 · CIFAR-100accuracy: 84.51 -> 84.62No clear effect
use TA instead of brute-force RAWide-ResNet-28-10 · Occluded CIFAR-10accuracy: 95.52 -> 95.72Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy