Overview

Learning to Optimize studies how data can improve the algorithms and decisions used to solve optimization problems, especially when similar problems recur or standard solvers are costly. Approaches include meta-learning and reinforcement learning for update rules and optimizer selection, unrolled or differentiable methods such as learned ADMM, and learned heuristics or warm starts for mixed-integer and other solvers. These methods are used in settings ranging from neural-network training and black-box search to scheduling, power systems, and model predictive control, with work increasingly combining fast learned predictions with exact optimization or explicit feasibility checks. A central tension is whether methods trained on particular instances can generalize reliably to new problems while preserving solution quality, constraint satisfaction, and the computational guarantees of classical solvers.

Latest papers

152 papers16 papers in the last four weeks, down 30% on the four weeks before. 0.1% of all new papers.

May 29, 2024

MEGO: Learning Mixture-of-Experts for General-Purpose Binary Optimization

Discrete optimization is ubiquitous in science and engineering. The vast array of existing discrete optimization problems, coupled with the continuous emergence of new ones, necessitates off-the-shelf optimizers capable of generating high-quality solutions for a large variety of optimization problems. This article introduces MEGO, a novel general-purpose neural optimizer for binary optimization under the black-box setting, intended for broad applicability across diverse binary optimization problem classes with minimal problem-specific customization. MEGO comprises a mixture-of-experts trained without domain knowledge. When presented with a new problem instance to solve, it employs a routing policy to dynamically activate the most relevant expert models to generate high-quality solutions. The strong generalization capability of MEGO is demonstrated on six problem classes from different disciplines, including classic problems and real-world applications. Trained solely on classic problems, MEGO effectively generalizes to unseen and complex real-world problem classes, significantly outperforming widely-used general-purpose optimizers in both solution quality and efficiency. Furthermore, MEGO provides a computational approach for quantifying similarity between optimization problems and classifying them, which is fundamentally different from the conventional analysis-based problem classification.
Oct 27, 2023

LOOP-MAC: Learning-Based Multi-Agent Coordination with Guaranteed Local Feasibility in a Virtual Power Plant

By aggregating distributed energy resources (DERs) into a single dispatchable resource, virtual power plants (VPPs) enable the grid to use the flexibility of numerous small-scale DERs. Delivering an aggregate VPP commitment requires multi-agent coordination, where agents manage local DER models and operating constraints while jointly meeting a power commitment within limited communication rounds. Classical distributed optimization is computationally intensive, requiring iterative local optimization and repeated communication to reach consensus. Learning-based methods can accelerate coordination but often lack guarantees of local feasibility and accurate commitment tracking. This paper proposes LOOP-MAC (Learning to Optimize the Optimization Process for Multi-Agent Coordination), a fully decentralized learning-based framework that ensures local feasibility by construction while efficiently reducing the aggregate commitment gap. LOOP-MAC employs a two-stage decision process. Learned coordination uses neural optimizers equipped with a gauge layer to produce locally feasible schedules through a few rounds of message exchange. Local adjustment then rebalances each schedule toward feasible flexibility extremes using a commitment-gap estimate inferred from messages already received. The adjusted schedule remains locally feasible as a convex combination of feasible schedules, and the adjustment requires no additional communication round. On a 100-agent VPP, five communication rounds followed by local adjustment reduce the mean commitment gap from 5.1% to 0.75% of the agents' aggregate absolute power and the optimality gap from 3.93% to 3.07% relative to learned coordination alone. Per-agent online computation takes approximately 1.1 ms, including the adjustment.