Skip to main content
Cornell University
We gratefully acknowledge support from the Simons Foundation, member institutions, and all contributors. Donate
arxiv logo > math-ph > arXiv:1810.12160

Help | Advanced Search

arXiv logo
Cornell University Logo

quick links

  • Login
  • Help Pages
  • About

Mathematical Physics

arXiv:1810.12160 (math-ph)
[Submitted on 29 Oct 2018]

Title:The Relativistic Hopfield network: rigorous results

Authors:Elena Agliari, Adriano Barra, Matteo Notarnicola
View a PDF of the paper titled The Relativistic Hopfield network: rigorous results, by Elena Agliari and 2 other authors
View PDF
Abstract:The relativistic Hopfield model constitutes a generalization of the standard Hopfield model that is derived by the formal analogy between the statistical-mechanic framework embedding neural networks and the Lagrangian mechanics describing a fictitious single-particle motion in the space of the tuneable parameters of the network itself. In this analogy the cost-function of the Hopfield model plays as the standard kinetic-energy term and its related Mattis overlap (naturally bounded by one) plays as the velocity. The Hamiltonian of the relativisitc model, once Taylor-expanded, results in a P-spin series with alternate signs: the attractive contributions enhance the information-storage capabilities of the network, while the repulsive contributions allow for an easier unlearning of spurious states, conferring overall more robustness to the system as a whole. Here we do not deepen the information processing skills of this generalized Hopfield network, rather we focus on its statistical mechanical foundation. In particular, relying on Guerra's interpolation techniques, we prove the existence of the infinite volume limit for the model free-energy and we give its explicit expression in terms of the Mattis overlaps. By extremizing the free energy over the latter we get the generalized self-consistent equations for these overlaps, as well as a picture of criticality that is further corroborated by a fluctuation analysis. These findings are in full agreement with the available previous results.
Comments: 11 pages, 1 figure
Subjects: Mathematical Physics (math-ph); Disordered Systems and Neural Networks (cond-mat.dis-nn)
Report number: Roma01.Math
Cite as: arXiv:1810.12160 [math-ph]
  (or arXiv:1810.12160v1 [math-ph] for this version)
  https://doi.org/10.48550/arXiv.1810.12160
arXiv-issued DOI via DataCite
Related DOI: https://doi.org/10.1063/1.5077060
DOI(s) linking to related resources

Submission history

From: Elena Agliari [view email]
[v1] Mon, 29 Oct 2018 14:40:20 UTC (98 KB)
Full-text links:

Access Paper:

    View a PDF of the paper titled The Relativistic Hopfield network: rigorous results, by Elena Agliari and 2 other authors
  • View PDF
  • TeX Source
  • Other Formats
view license
Current browse context:
math-ph
< prev   |   next >
new | recent | 2018-10
Change to browse by:
cond-mat
cond-mat.dis-nn
math
math.MP

References & Citations

  • NASA ADS
  • Google Scholar
  • Semantic Scholar
a export BibTeX citation Loading...

BibTeX formatted citation

×
Data provided by:

Bookmark

BibSonomy logo Reddit logo

Bibliographic and Citation Tools

Bibliographic Explorer (What is the Explorer?)
Connected Papers (What is Connected Papers?)
Litmaps (What is Litmaps?)
scite Smart Citations (What are Smart Citations?)

Code, Data and Media Associated with this Article

alphaXiv (What is alphaXiv?)
CatalyzeX Code Finder for Papers (What is CatalyzeX?)
DagsHub (What is DagsHub?)
Gotit.pub (What is GotitPub?)
Hugging Face (What is Huggingface?)
Papers with Code (What is Papers with Code?)
ScienceCast (What is ScienceCast?)

Demos

Replicate (What is Replicate?)
Hugging Face Spaces (What is Spaces?)
TXYZ.AI (What is TXYZ.AI?)

Recommenders and Search Tools

Influence Flower (What are Influence Flowers?)
CORE Recommender (What is CORE?)
  • Author
  • Venue
  • Institution
  • Topic

arXivLabs: experimental projects with community collaborators

arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.

Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.

Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.

Which authors of this paper are endorsers? | Disable MathJax (What is MathJax?)
  • About
  • Help
  • contact arXivClick here to contact arXiv Contact
  • subscribe to arXiv mailingsClick here to subscribe Subscribe
  • Copyright
  • Privacy Policy
  • Web Accessibility Assistance
  • arXiv Operational Status
    Get status notifications via email or slack