A Neural Network Framework for Discovering Closed-form Solutions to Quadratic Programs with Linear Constraints

TL;DR

This paper introduces a neural network method that analytically derives exact closed-form solutions for quadratic programs with linear constraints, ensuring optimality and feasibility without training.

Contribution

The proposed approach analytically computes NN parameters from problem data to discover exact solutions, bypassing traditional training and guaranteeing accuracy across all solution regions.

Findings

Outperformed classical DNN in KKT optimality and feasibility.

Achieved competitive or better performance than Gurobi solver.

Solved large-scale energy planning problems in seconds.

Abstract

Deep neural networks (DNNs) have been used to model complex optimization problems in many applications, yet have difficulty guaranteeing solution optimality and feasibility, despite training on large datasets. Training a NN as a surrogate optimization solver amounts to estimating a global solution function that maps varying problem input parameters to the corresponding optimal solutions. Work in multiparametric programming (mp) has shown that solutions to quadratic programs (QP) are piece-wise linear functions of the parameters, and researchers have suggested leveraging this property to model mp-QP using NN with ReLU activation functions, which also exhibit piecewise linear behaviour. This paper proposes a NN modeling approach and learning algorithm that discovers the exact closed-form solution to QP with linear constraints, by analytically deriving NN model parameters directly from the problem coefficients without training. Whereas generic DNN cannot guarantee accuracy outside the training distribution, the closed-form NN model produces exact solutions for every discovered critical region of the solution function. To evaluate the closed-form NN model, it was applied to DC optimal power flow problems in electricity management. In terms of Karush-Kuhn-Tucker (KKT) optimality and feasibility of solutions, it outperformed a classically trained DNN and was competitive with, or outperformed, a commercial analytic solver (Gurobi) at far less computational cost. For a long-range energy planning problem, it was able to produce optimal and feasible solutions for millions of input parameters within seconds.