Akshay Agrawal, Shane Barratt, Stephen Boyd, Enzo Busseti, Walaa M. Moursi, Differentiating through a cone program
Full Text: PDF
DOI: 10.23952/jano.1.2019.2.02
Volume 1, Issue 2, 31 August 2019, Pages 107-115
Abstract. We consider the problem of efficiently computing the derivative of the solution map of a convex cone program, when it exists. We do this by implicitly differentiating the residual map for its homogeneous self-dual embedding, and solving the linear systems of equations required using an iterative method. This allows us to efficiently compute the derivative operator, and its adjoint, evaluated at a vector. These correspond to computing an approximate new solution, given a perturbation to the cone program coefficients (i.e., perturbation analysis), and to computing the gradient of a function of the solution with respect to the coefficients. Our method scales to large problems, with numbers of coefficients in the millions. We present an open-source Python implementation of our method that solves a cone program and returns the derivative and its adjoint as abstract linear maps; our implementation can be easily integrated into software systems for automatic differentiation.
How to Cite this Article:
Akshay Agrawal, Shane Barratt, Stephen Boyd, Enzo Busseti, Walaa M. Moursi, Differentiating through a cone program, J. Appl. Numer. Optim. 1 (2019), 107-115.