Ravi Tandon

Information Theoretic Limits of Data Shuffling for Distributed Learning

2016

Data shuffling is one of the fundamental building blocks for distributed learning algorithms, that increases the statistical gain for each step of the learning process. In each iteration, different shuffled data points are assigned by …

On the Capacity of Secure Distributed Matrix Multiplication

2018 · arXiv (Cornell University)

Matrix multiplication is one of the key operations in various engineering applications. Outsourcing large-scale matrix multiplication tasks to multiple distributed servers or cloud is desirable to speed up computation. However, security becomes an issue when …

Privacy Amplification for Federated Learning via User Sampling and\n Wireless Aggregation

2021 · arXiv (Cornell University)

In this paper, we study the problem of federated learning over a wireless\nchannel with user sampling, modeled by a Gaussian multiple access channel,\nsubject to central and local differential privacy (DP/LDP) constraints. It has\nbeen shown that …

Adaptive Privacy for Differentially Private Causal Graph Discovery

2024

Causal Graph Discovery (CGD) enables the estimation of directed acyclic graph (DAG) that represents the joint probability distribution of observational data. To estimate DAGs, typical constraint-based CGD algorithms run a sequence of conditional independence (CI) …