Data-Copying in Generative Models: A Formal Framework
Abstract
There has been some recent interest in detecting and addressing memorization of training data by deep neural networks. A formal framework for memorization in generative models, called “data-copying” was proposed by Meehan et. al (2020). We build upon their work to show that their framework may fail to detect certain kinds of blatant memorization. Motivated by this and the theory of non-parametric methods, we provide an alternative definition of data-copying that applies more locally. We provide a method to detect data-copying, and provably show that it works with high probability when enough data is available. We also provide lower bounds that characterize the sample requirement for reliable detection.
Cite
Text
Bhattacharjee et al. "Data-Copying in Generative Models: A Formal Framework." International Conference on Machine Learning, 2023.Markdown
[Bhattacharjee et al. "Data-Copying in Generative Models: A Formal Framework." International Conference on Machine Learning, 2023.](https://mlanthology.org/icml/2023/bhattacharjee2023icml-datacopying/)BibTeX
@inproceedings{bhattacharjee2023icml-datacopying,
title = {{Data-Copying in Generative Models: A Formal Framework}},
author = {Bhattacharjee, Robi and Dasgupta, Sanjoy and Chaudhuri, Kamalika},
booktitle = {International Conference on Machine Learning},
year = {2023},
pages = {2364-2396},
volume = {202},
url = {https://mlanthology.org/icml/2023/bhattacharjee2023icml-datacopying/}
}