Salvato in:
Dettagli Bibliografici
Autori principali: Maliakal, Gabriel, Alkhouri, Ismail, Velasquez, Alvaro, Alessio, Adam M, Ravishankar, Saiprasad
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:https://arxiv.org/abs/2505.13405
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866915345559715840
author Maliakal, Gabriel
Alkhouri, Ismail
Velasquez, Alvaro
Alessio, Adam M
Ravishankar, Saiprasad
author_facet Maliakal, Gabriel
Alkhouri, Ismail
Velasquez, Alvaro
Alessio, Adam M
Ravishankar, Saiprasad
contents The Maximum Cut (MaxCut) problem is NP-Complete, and obtaining its optimal solution is NP-hard in the worst case. As a result, heuristic-based algorithms are commonly used, though their design often requires significant domain expertise. More recently, learning-based methods trained on large (un)labeled datasets have been proposed; however, these approaches often struggle with generalizability and scalability. A well-known approximation algorithm for MaxCut is the Goemans-Williamson (GW) algorithm, which relaxes the Quadratic Unconstrained Binary Optimization (QUBO) formulation into a semidefinite program (SDP). The GW algorithm then applies hyperplane rounding by uniformly sampling a random hyperplane to convert the SDP solution into binary node assignments. In this paper, we propose a training-data-free approach based on a non-episodic reinforcement learning formulation, in which an agent learns to select improved rounding hyperplanes that yield better cuts than those produced by the GW algorithm. By optimizing over a Markov Decision Process (MDP), our method consistently achieves better cuts across large-scale graphs with varying densities and degree distributions.
format Preprint
id arxiv_https___arxiv_org_abs_2505_13405
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle A Dataless Reinforcement Learning Approach to Rounding Hyperplane Optimization for Max-Cut
Maliakal, Gabriel
Alkhouri, Ismail
Velasquez, Alvaro
Alessio, Adam M
Ravishankar, Saiprasad
Machine Learning
The Maximum Cut (MaxCut) problem is NP-Complete, and obtaining its optimal solution is NP-hard in the worst case. As a result, heuristic-based algorithms are commonly used, though their design often requires significant domain expertise. More recently, learning-based methods trained on large (un)labeled datasets have been proposed; however, these approaches often struggle with generalizability and scalability. A well-known approximation algorithm for MaxCut is the Goemans-Williamson (GW) algorithm, which relaxes the Quadratic Unconstrained Binary Optimization (QUBO) formulation into a semidefinite program (SDP). The GW algorithm then applies hyperplane rounding by uniformly sampling a random hyperplane to convert the SDP solution into binary node assignments. In this paper, we propose a training-data-free approach based on a non-episodic reinforcement learning formulation, in which an agent learns to select improved rounding hyperplanes that yield better cuts than those produced by the GW algorithm. By optimizing over a Markov Decision Process (MDP), our method consistently achieves better cuts across large-scale graphs with varying densities and degree distributions.
title A Dataless Reinforcement Learning Approach to Rounding Hyperplane Optimization for Max-Cut
topic Machine Learning
url https://arxiv.org/abs/2505.13405