Data Augmentation Strategy

Data augmentation strategies aim to enhance the performance of machine learning models, particularly in scenarios with limited training data, by artificially expanding the dataset with modified versions of existing samples. Current research focuses on optimizing augmentation techniques for specific model architectures (e.g., convolutional neural networks, vision transformers) and tasks (e.g., image classification, object detection, sequential recommendation), often employing automated search algorithms or generative models to identify effective augmentation policies. These advancements are crucial for improving model robustness, accuracy, and efficiency across diverse applications, ranging from medical image analysis and industrial predictive maintenance to natural language processing and reinforcement learning.

Papers

August 20, 2024

A Closer Look at Data Augmentation Strategies for Finetuning-Based Low/Few-Shot Object Detection
Vladislav Li, Georgios Tsoumplekas, Ilias Siniosoglou, Vasileios Argyriou, Anastasios Lytos, Eleftherios Fountoukidis, Panagiotis Sarigiannidis
Glance Annotation Shot Object Detection Low Priority Data Augmentation Strategy Lightweight Object Detection

June 18, 2024

Class-specific Data Augmentation for Plant Stress Classification
Nasla Saleem, Aditya Balu, Talukder Zaki Jubery, Arti Singh, Asheesh K. Singh, Soumik Sarkar, Baskar Ganapathysubramanian
Data Augmentation Data Augmentation Strategy Plant Stress

May 27, 2024

GTA: Generative Trajectory Augmentation with Guidance for Offline Reinforcement Learning
Jaewoo Lee, Sujin Yun, Taeyoung Yun, Jinkyoo Park
Data Augmentation Offline Reinforcement Learning Human Guidance Generative Data Augmentation Data Augmentation Strategy Trajectory Augmentation

May 5, 2024

You Only Need Half: Boosting Data Augmentation by Using Partial Content
Juntao Hu, Yuan Wu
Adversarial Attack Data Augmentation Data Augmentation Technique Novel Data Augmentation Data Augmentation Strategy Image Data Augmentation Missing Half

November 15, 2023

Strategic Data Augmentation with CTGAN for Smart Manufacturing: Enhancing Machine Learning Predictions of Paper Breaks in Pulp-and-Paper Production
Hamed Khosravi, Sarah Farhadpour, Manikanta Grandhi, Ahmed Shoyeb Raihan, Srinjoy Das, Imtiaz Ahmed
Machine Learning Generative Adversarial Network Predictive Model Rare Event Smart Manufacturing Predictive Maintenance Data Augmentation Strategy

September 18, 2023

Replication: Contrastive Learning and Data Augmentation in Traffic Classification Using a Flowpic Input Representation
Alessandro Finamore, Chao Wang, Jonatan Krolikowski, Jose M. Navarro, Fuxing Chen, Dario Rossi
Deep Learning Contrastive Learning Data Augmentation Traffic Classification Serial Reproduction Data Augmentation Strategy

September 13, 2023

The effect of data augmentation and 3D-CNN depth on Alzheimer's Disease detection
Rosanna Turrisi, Alessandro Verri, Annalisa Barla
Machine Learning Data Augmentation Mixed Effect Alzheimer'S Disease Large Depth Disease Detection Data Augmentation Method Data Augmentation Strategy

September 7, 2023

Understanding Data Augmentation from a Robustness Perspective
Zhendong Liu, Jie Zhang, Qiangqiang He, Chongjun Wang
Data Augmentation Model Robustness Augmentation Technique Data Augmentation Strategy Player Interaction

May 30, 2023

Joint Optimization of Class-Specific Training- and Test-Time Data Augmentation in Segmentation
Zeju Li, Konstantinos Kamnitsas, Qi Dou, Chen Qin, Ben Glocker
Data Augmentation Segmentation Based Approach Joint Optimization Test Time Augmentation Data Augmentation Strategy

May 29, 2023

Data Augmentation for Low-Resource Keyphrase Generation
Krishna Garg, Jishnu Ray Chowdhury, Cornelia Caragea
Data Augmentation Keyphrase Generation Author Assigned Keyphrases Data Augmentation Strategy

April 27, 2023

NAP at SemEval-2023 Task 3: Is Less Really More? (Back-)Translation as Data Augmentation Strategies for Detecting Persuasion Techniques
Neele Falk, Annerose Eichel, Prisca Piccirilli
SemEval 2022 Task Back Translation Multilingual Transformer Multilingual Scenario Data Augmentation Strategy Persuasion Strategy Rest RESTAD NAP Persuasion Technique Detection

March 26, 2023

Analyzing Effects of Mixed Sample Data Augmentation on Model Interpretability
Soyoun Won, Sung-Ho Bae, Seong Tae Kim
Data Augmentation Model Interpretability Data Augmentation Strategy Mixed Sample Data Augmentation

February 27, 2023

A Comparison of Speech Data Augmentation Methods Using S3PRL Toolkit
Mina Huh, Ruchira Ray, Corey Karnei
Data Augmentation Automatic Speech Recognition Consistent Comparison Augmentation Technique Phoneme Recognition Data Augmentation Strategy Speech Data Augmentation

February 14, 2023

Adding Instructions during Pretraining: Effective Way of Controlling Toxicity in Language Models
Shrimai Prabhumoye, Mostofa Patwary, Mohammad Shoeybi, Bryan Catanzaro
Large Language Model Language Model Natural Language Processing Human Instruction Non Toxic Effective Method Natural Language Processing Benchmark Data Augmentation Strategy

November 22, 2022

Improving Crowded Object Detection via Copy-Paste
Jiangfan Deng, Dewen Fan, Xiaosong Qiu, Feng Zhou
Data Augmentation Object Detection Data Augmentation Strategy Copy Paste Consensus Learning Crowded Pedestrian Detection

November 10, 2022

MixUp-MIL: Novel Data Augmentation for Multiple Instance Learning and a Study on Thyroid Cancer Diagnosis
Michael Gadermayr, Lukas Koller, Maximilian Tschuchnig, Lea Maria Stangassinger, Christina Kreutzer, Sebastien Couillard-Despres, Gertie Janneke Oostingh, Anton Hittmair
Data Augmentation Study Feature Multiple Instance Learning Whole Slide Image Novel Data Augmentation Thyroid Cancer Data Augmentation Strategy Local Mixup

August 15, 2022

Syntax-driven Data Augmentation for Named Entity Recognition
Arie Pratama Sutiono, Gus Hahn-Powell
Data Augmentation Entity Recognition Token Level Text Augmentation Data Augmentation Strategy

July 16, 2022

On the Importance of Hyperparameters and Data Augmentation for Self-Supervised Learning
Diane Wagner, Fabio Ferreira, Danny Stoll, Robin Tibor Schirrmeister, Samuel Müller, Frank Hutter
Data Augmentation Self Supervised Learning Importance Aware Related Hyperparameters Data Augmentation Strategy Data Augmentation Algorithm

April 9, 2022

Data Augmentation for Electrocardiograms
Aniruddh Raghu, Divya Shanmugam, Eugene Pomerantsev, John Guttag, Collin M. Stultz
Data Augmentation Electrocardiogram Classification Data Augmentation Strategy Training Data Augmentation

March 26, 2022

Data Augmentation Strategies for Improving Sequential Recommender Systems
Joo-yeong Song, Bongwon Suh
Deep Learning Data Augmentation Sequential Recommender System Data Augmentation Strategy

Data Augmentation Strategy

Papers

A Closer Look at Data Augmentation Strategies for Finetuning-Based Low/Few-Shot Object Detection

Class-specific Data Augmentation for Plant Stress Classification

GTA: Generative Trajectory Augmentation with Guidance for Offline Reinforcement Learning

You Only Need Half: Boosting Data Augmentation by Using Partial Content

Strategic Data Augmentation with CTGAN for Smart Manufacturing: Enhancing Machine Learning Predictions of Paper Breaks in Pulp-and-Paper Production

Replication: Contrastive Learning and Data Augmentation in Traffic Classification Using a Flowpic Input Representation

The effect of data augmentation and 3D-CNN depth on Alzheimer's Disease detection

Understanding Data Augmentation from a Robustness Perspective

Joint Optimization of Class-Specific Training- and Test-Time Data Augmentation in Segmentation

Data Augmentation for Low-Resource Keyphrase Generation

NAP at SemEval-2023 Task 3: Is Less Really More? (Back-)Translation as Data Augmentation Strategies for Detecting Persuasion Techniques

Analyzing Effects of Mixed Sample Data Augmentation on Model Interpretability

A Comparison of Speech Data Augmentation Methods Using S3PRL Toolkit

Adding Instructions during Pretraining: Effective Way of Controlling Toxicity in Language Models

Improving Crowded Object Detection via Copy-Paste

MixUp-MIL: Novel Data Augmentation for Multiple Instance Learning and a Study on Thyroid Cancer Diagnosis

Syntax-driven Data Augmentation for Named Entity Recognition

On the Importance of Hyperparameters and Data Augmentation for Self-Supervised Learning

Data Augmentation for Electrocardiograms

Data Augmentation Strategies for Improving Sequential Recommender Systems