Training Data

Training data is crucial for machine learning model development, with current research focusing on improving data quality, efficiency, and mitigating biases. Active areas include generating synthetic data to address scarcity or privacy concerns, developing algorithms to optimize data selection and usage (e.g., self-paced learning, active learning), and mitigating issues like data contamination and imbalance through techniques such as data augmentation, selective parameter merging, and novel loss functions. The quality and characteristics of training data significantly impact model performance, generalization, and robustness, influencing various applications from natural language processing and image recognition to scientific computing and medical diagnosis.

Papers

January 31, 2023

January 30, 2023

Extracting Training Data from Diffusion Models
Nicholas Carlini, Jamie Hayes, Milad Nasr, Matthew Jagielski, Vikash Sehwag, Florian Tramèr, Borja Balle, Daphne Ippolito, Eric Wallace
Diffusion Model Training Data GAN Model Synthetic Image Image Diffusion Model Generative Prior Privacy Preserving Training

January 25, 2023

Backward Compatibility During Data Updates by Weight Interpolation
Raphael Schumann, Elman Mansimov, Yi-An Lai, Nikolaos Pappas, Xibin Gao, Yi Zhang
Training Data Model Prediction Incremental Data Software Defect Weight Interpolation

January 23, 2023

January 20, 2023

Split Ways: Privacy-Preserving Training of Encrypted Data Using Split Learning
Tanveer Khan, Khoa Nguyen, Antonis Michalas
Training Data Homomorphic Encryption Split Learning PRIvacy Leakage Collaborative Learning Split and Fit Training Data Privacy

January 18, 2023

Targeted Image Reconstruction by Sampling Pre-trained Diffusion Model
Jiageng Zheng
Training Data Pre Trained Diffusion Model Neural Network Model Open Sampling Model Inversion Attack Target Distribution

January 16, 2023

Learning from Very Little Data: On the Value of Landscape Analysis for Predicting Software Project Health
Andre Lustosa, Tim Menzies
LeArning Abstract Training Data Net Present Value Prediction Error Landscape Analysis Code Metric Less Data

January 14, 2023

Poisoning Attacks and Defenses in Federated Learning: A Survey
Subhash Sagar, Chang-Sun Li, Seng W. Loke, Jinho Choi
Timely Survey Training Data Poisoning Attack Security Threat

January 8, 2023

Prognosis and Treatment Prediction of Type-2 Diabetes Using Deep Neural Network and Machine Learning Classifiers
Md. Kowsher, Mahbuba Yesmin Turaba, Tanvir Sajed, M M Mahabubur Rahman
Neural Network Deep Neural Network Training Data Machine Learning Classifier Prognosis Prediction Type 2 Diabetes Accurate Predictive Model Treatment Outcome Prediction

January 5, 2023

January 3, 2023

Data Valuation Without Training of a Model
Nohyun Ki, Hoyong Choi, Hye Won Chung
Deep Learning Training Data Full Model Data Valuation Multi Scoring

January 2, 2023

Training Differentially Private Graph Neural Networks with Random Walk Sampling
Morgane Ayle, Jan Schuchardt, Lukas Gosch, Daniel Zügner, Stephan Günnemann
Training Data Graph Structured Data Privacy Guarantee Large Graph Private Stochastic Gradient Descent Shallow Network Private Graph Neural Network

January 1, 2023

Pruning Before Training May Improve Generalization, Provably
Hongru Yang, Yingbin Liang, Xiaojie Guo, Lingfei Wu, Zhangyang Wang
Strong Generalization Training Data Gradient Descent Generalization Performance Edge Pruning Generalization Capability Pruning Framework

December 31, 2022

Assessment of creditworthiness models privacy-preserving training with synthetic data
Ricardo Muñoz-Cancino, Cristián Bravo, Sebastián A. Ríos, Manuel Graña
Training Data Synthetic Data Direct Assessment Credit Risk Credit Scoring Model

December 27, 2022