Machine-learning-assisted Monte Carlo fails at sampling computationally hard problems - Archive ouverte HAL
Article Dans Une Revue Machine Learning: Science and Technology Année : 2023

Machine-learning-assisted Monte Carlo fails at sampling computationally hard problems

Résumé

Abstract Several strategies have been recently proposed in order to improve Monte Carlo sampling efficiency using machine learning tools. Here, we challenge these methods by considering a class of problems that are known to be exponentially hard to sample using conventional local Monte Carlo at low enough temperatures. In particular, we study the antiferromagnetic Potts model on a random graph, which reduces to the coloring of random graphs at zero temperature. We test several machine-learning-assisted Monte Carlo approaches, and we find that they all fail. Our work thus provides good benchmarks for future proposals for smart sampling algorithms.
Fichier principal
Vignette du fichier
Ciarella_2023_Mach._Learn.__Sci._Technol._4_010501.pdf (19.11 Mo) Télécharger le fichier
Origine Fichiers éditeurs autorisés sur une archive ouverte
Licence

Dates et versions

hal-04028482 , version 1 (10-09-2024)

Licence

Identifiants

Citer

Simone Ciarella, Jeanne Trinquier, Martin Weigt, Francesco Zamponi. Machine-learning-assisted Monte Carlo fails at sampling computationally hard problems. Machine Learning: Science and Technology, 2023, 4 (1), pp.010501. ⟨10.1088/2632-2153/acbe91⟩. ⟨hal-04028482⟩
24 Consultations
0 Téléchargements

Altmetric

Partager

More