End-to-End Constrained Optimization Learning: A Survey

TL;DR

This survey reviews recent developments in integrating machine learning with constrained optimization, highlighting hybrid methods that aim to predict approximate solutions and enable logical inference for combinatorial problems.

Contribution

It provides a comprehensive overview of recent approaches combining machine learning with constrained optimization and combinatorial solvers.

Findings

Hybrid methods can predict approximate solutions efficiently.

Integration of ML with optimization enables structural logical inference.

Recent advancements show promising directions for future research.

Abstract

This paper surveys the recent attempts at leveraging machine learning to solve constrained optimization problems. It focuses on surveying the work on integrating combinatorial solvers and optimization methods with machine learning architectures. These approaches hold the promise to develop new hybrid machine learning and optimization methods to predict fast, approximate, solutions to combinatorial problems and to enable structural logical inference. This paper presents a conceptual review of the recent advancements in this emerging area.