Tight Low Degree Hardness for Optimizing Pure Spherical Spin Glasses

Fuente: arXiv
Saved in:
Bibliographic Details
Main Author: Sellke, Mark
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915231511347200
author Sellke, Mark
author_facet Sellke, Mark
contents We prove constant degree polynomial algorithms cannot optimize pure spherical $p$-spin Hamiltonians beyond the algorithmic threshold $\mathsf{ALG}(p)=2\sqrt{\frac{p-1}{p}}$. The proof goes by transforming any hypothetical such algorithm into a Lipschitz one, for which hardness was shown previously by the author and B. Huang.
format Preprint
id arxiv_https___arxiv_org_abs_2504_04632
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Tight Low Degree Hardness for Optimizing Pure Spherical Spin Glasses
Sellke, Mark
Probability
Disordered Systems and Neural Networks
Data Structures and Algorithms
Mathematical Physics
We prove constant degree polynomial algorithms cannot optimize pure spherical $p$-spin Hamiltonians beyond the algorithmic threshold $\mathsf{ALG}(p)=2\sqrt{\frac{p-1}{p}}$. The proof goes by transforming any hypothetical such algorithm into a Lipschitz one, for which hardness was shown previously by the author and B. Huang.
title Tight Low Degree Hardness for Optimizing Pure Spherical Spin Glasses
topic Probability
Disordered Systems and Neural Networks
Data Structures and Algorithms
Mathematical Physics
url https://arxiv.org/abs/2504.04632