Gradient boosting for convex cone predict and optimize problems

Abstract

Prediction models are traditionally optimized independently from decision-based optimization. Conversely, a ‘smart predict then optimize’ (SPO) framework optimizes prediction models to minimize downstream decision regret. In this paper we present dboost, the first general purpose implementation of smart gradient boosting for ‘predict, then optimize’ problems. The framework supports convex quadratic cone programming and gradient boosting is performed by implicit differentiation of a custom fixed-point mapping. Experiments comparing with state-of-the-art SPO methods show that dboost can further reduce out-of-sample decision regret.