Numerical studies of a one-dimensional 3-spin spin-glass model with long-range interactions
Derek Larson, Helmut G. Katzgraber, M. A. Moore, A. P. Young

TL;DR
This study investigates whether the finite-temperature glass transition observed in mean-field p-spin models persists in non-mean-field regimes by using a 3-spin model with tunable long-range interactions, revealing a loss of transition deep in the non-mean-field regime.
Contribution
It provides the first systematic numerical analysis of a 3-spin model with long-range interactions across different effective dimensions, connecting mean-field and non-mean-field behaviors.
Findings
Finite-temperature transition is absent deep in the non-mean-field regime.
Transition persists in the mean-field regime, consistent with theoretical predictions.
An apparent transition appears in the non-mean-field region, possibly due to finite-size effects.
Abstract
We study a p-spin spin-glass model to understand if the finite-temperature glass transition found in the mean-field regime of p-spin models, and used to model the behavior of structural glasses, persists in the non-mean-field regime. By using a 3-spin spin-glass model with long-range power-law diluted interactions we are able to continuously tune the (effective) space dimension via the exponent of the interactions. Monte Carlo simulations of the spin-glass susceptibility and the two-point finite-size correlation length show that deep in the non-mean-field regime the finite-temperature transition is lost, whereas this is not the case in the mean-field regime, in agreement with the prediction of Moore and Drossel [Phys. Rev. Lett. 89, 217202 (2002)] that 3-spin models are in the same universality class as an Ising spin glass in a magnetic field. However, slightly in the non-mean-field…
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
