Treballs Fi de GrauEnginyeria Informàtica i Matemàtiques

Hopfield Network and beyond

  • Identification data

    Identifier:  TFG:9575
    Handle: https://hdl.handle.net/20.500.11797/TFG9575
    Authors:  Rios Blanch, Guifré de los
    Abstract:
    This work studies Hopfield networks as models of associative memory, both in their classical and modern versions. The mathematical and computational foundations are explored, and several experiments are programmed to analyze the ability to recover binary digits, both in low and high resolution. Different learning rules (Hebb and pseudoinverse) and recovery dynamics (synchronous and asynchronous) are compared, analyzing the results obtained as a function of noise, resolution and iterations.

  • Others:

    Access rights: info:eu-repo/semantics/openAccess
    Education area(s): Enginyeria Matemàtica i Física
    Department: Enginyeria Informàtica i Matemàtiques
    Entity: Universitat Rovira i Virgili (URV)
    Confidenciality: No
    Subject: Xarxes neuronals (informàtica)
    Project director: Arenas Moreno, Alejandro
    Work's public defense date: 2025
    Creation date in repository: 2026-07-06
    Academic year: 2024-2025
    Student: Rios Blanch, Guifré de los

  • Keywords:

    Associative memory
    Hopfield networks
    Energy minimization
    Mathematical Engineering and Physics

  • Documents:

    Memòria

  • Cerca a google

    Search to google scholar