Gautam Kamath

Differentially Private Algorithms for Learning Mixtures of Separated\n Gaussians

2019 · arXiv (Cornell University)

Learning the parameters of Gaussian mixture models is a fundamental and\nwidely studied problem with numerous applications. In this work, we give new\nalgorithms for learning the parameters of a high-dimensional, well separated,\nGaussian mixture model subject to …

Indiscriminate Data Poisoning Attacks on Pre-trained Feature Extractors

2024

Machine learning models have achieved great success in supervised learning tasks for end-to-end training, which requires a large amount of labeled data that is not always feasible. Recently, many practitioners have shifted to self-supervised learning …

Differentially Private Fine-tuning of Language Models

2024 · Journal of Privacy and Confidentiality

We give simpler, sparser, and faster algorithms for differentially private fine-tuning of large-scale pre-trained language models, which achieve the state-of-the-art privacy versus utility tradeoffs on many standard NLP tasks. We propose a meta-framework for this …