Paper accepted at ESORICS'24


Attesting Distributional Properties of Training Data for Machine Learning by Vasisht Duddu, Anudeep Das, Nora Khayata , Hossein Yalame , Thomas Schneider , and N. Asokan was accepted at the top conference 29th European Symposium on Research in Computer Security (ESORICS) 2024.

This paper is a collaboration with Vasisht Duddu, Anudeep Das, and N. Asokan from the University of Waterloo.

It proposes the notion of property attestation using Machine Learning, cryptographic techniques, and a mix of both to demonstrate relevant distributional properties of training data in conjunction with the trained model without revealing the data.