Iterative Sketching for Secure Coded Regression
This work addresses the need for secure and straggler-resilient distributed computing in linear regression applications, but it appears incremental as it builds on existing coded computing frameworks.
The paper tackles the problem of speeding up distributed linear regression by proposing a method that uses randomized basis rotation and subsampling to secure information and reduce dimensionality, resulting in a distributive iterative stochastic approach for matrix compression and steepest descent.
Linear regression is a fundamental and primitive problem in supervised machine learning, with applications ranging from epidemiology to finance. In this work, we propose methods for speeding up distributed linear regression. We do so by leveraging randomized techniques, while also ensuring security and straggler resiliency in asynchronous distributed computing systems. Specifically, we randomly rotate the basis of the system of equations and then subsample blocks, to simultaneously secure the information and reduce the dimension of the regression problem. In our setup, the basis rotation corresponds to an encoded encryption in an approximate gradient coding scheme, and the subsampling corresponds to the responses of the non-straggling servers in the centralized coded computing framework. This results in a distributive iterative stochastic approach for matrix compression and steepest descent.