A Primal-Dual Operator Splitting Method for Conic Optimization

Working Draft.

• Current version

• Related paper

• Source code

We develop a simple operator splitting method for solving a primal conic optimization problem; we show that the iterates also solve the dual problem. The resulting algorithm is very simple to describe and implement and yields solutions of modest accuracy in competitive times. Several versions of the algorithm are amenable to parallelization, either via distributed linear algebra or GPU-accelerated matrix-vector multiplication. We provide a simple, single-threaded C implementation for reference.