SOTAVerified

Data Augmentation

Data augmentation involves techniques used for increasing the amount of data, based on different modifications, to expand the amount of examples in the original dataset. Data augmentation not only helps to grow the dataset but it also increases the diversity of the dataset. When training machine learning models, data augmentation acts as a regularizer and helps to avoid overfitting.

Data augmentation techniques have been found useful in domains like NLP and computer vision. In computer vision, transformations like cropping, flipping, and rotation are used. In NLP, data augmentation techniques can include swapping, deletion, random insertion, among others.

Further readings:

( Image credit: Albumentations )

Papers

Recently AddedMost HypedMost ActiveNeeds VerificationMost Verified

Showing 6751–6775 of 8378 papers

TitleStatusHype
On the Performance of Convolutional Neural Networks under High and Low Frequency Information—0
Acoustic Scene Classification with Squeeze-Excitation Residual Networks—0
On the Pitfalls of Learning with Limited Data: A Facial Expression Recognition Case Study—0
On the Reproducibility of Neural Network Predictions—0
On the Robustness of Human-Object Interaction Detection against Distribution Shift—0
On the Robustness of Speech Emotion Recognition for Human-Robot Interaction with Deep Neural Networks—0
On the Role of Supervision in Unsupervised Constituency Parsing—0
On the (Un-)Avoidability of Adversarial Examples—0
On the Usability of Transformers-based models for a French Question-Answering task—0
On the Usefulness of Synthetic Tabular Data Generation—0
On the Way to LLM Personalization: Learning to Remember User Conversations—0
ON-TRAC Consortium for End-to-End and Simultaneous Speech Translation Challenge Tasks at IWSLT 2020—0
On Training Sketch Recognizers for New Domains—0
On Using SpecAugment for End-to-End Speech Translation—0
OoMMix: Out-of-manifold Regularization in Contextual Embedding Space for Text Classification—0
OOWL500: Overcoming Dataset Collection Bias in the Wild—0
Open data for Moroccan license plates for OCR applications : data collection, labeling, and model construction—0
Open Set RF Fingerprinting using Generative Outlier Augmentation—0
Open Source Dataset and Machine Learning Techniques for Automatic Recognition of Historical Graffiti—0
Operationalizing Convolutional Neural Network Architectures for Prohibited Object Detection in X-Ray Imagery—0
OptGAN: Optimizing and Interpreting the Latent Space of the Conditional Text-to-Image GANs—0
Optical Character Recognition using Convolutional Neural Networks for Ashokan Brahmi Inscriptions—0
Optical Flow Techniques for Facial Expression Analysis -- a Practical Evaluation Study—0
Optimal Layer Selection for Latent Data Augmentation—0
Optimal Resource Allocation for Serverless Queries—0
Show:102550
← PrevPage 271 of 336Next →
All datasetsImageNetCIFAR-10GA1457

Benchmark Results

#ModelMetricClaimedVerifiedStatus
1DeiT-B (+MixPro)Accuracy (%)82.9—Unverified
2ResNet-200 (DeepAA)Accuracy (%)81.32—Unverified
3DeiT-S (+MixPro)Accuracy (%)81.3—Unverified
4ResNet-200 (Fast AA)Accuracy (%)80.6—Unverified
5ResNet-200 (UA)Accuracy (%)80.4—Unverified
6ResNet-200 (AA)Accuracy (%)80—Unverified
7ResNet-50 (DeepAA)Accuracy (%)78.3—Unverified
8ResNet-50 (TA wide)Accuracy (%)78.07—Unverified
9ResNet-50 (LoRot-E)Accuracy (%)77.72—Unverified
10ResNet-50 (LoRot-I)Accuracy (%)77.71—Unverified
#ModelMetricClaimedVerifiedStatus
1WideResNet-40-2 (Faster AA)Percentage error3.7—Unverified
2Shake-Shake (26 2×32d) (Faster AA)Percentage error2.7—Unverified
3WideResNet-28-10 (Faster AA)Percentage error2.6—Unverified
4Shake-Shake (26 2×112d) (Faster AA)Percentage error2—Unverified
5Shake-Shake (26 2×96d) (Faster AA)Percentage error2—Unverified
#ModelMetricClaimedVerifiedStatus
1DiffAugClassification Accuracy92.7—Unverified
2PaCMAPClassification Accuracy85.3—Unverified
3hNNEClassification Accuracy77.4—Unverified
4TopoAEClassification Accuracy74.6—Unverified