We consider dense, associative neural-networks trained with no supervision and we investigate their computational capabilities analytically, via statistical-mechanics tools, and numerically, via Monte Carlo simulations. In particular, we obtain a phase diagram summarizing their performance as a function of the control parameters (e.g. quality and quantity of the training dataset, network storage, noise) that is valid in the limit of large network size and structureless datasets. Moreover, we establish a bridge between macroscopic observables standardly used in statistical mechanics and loss functions typically used in the machine learning. As technical remarks, from the analytical side, we extend Guerra's interpolation to tackle the non-Gaussian distributions involved in the post-synaptic potentials while, from the computational counterpart, we insert Plefka's approximation in the Monte Carlo scheme, to speed up the evaluation of the synaptic tensor, overall obtaining a novel and broad approach to investigate unsupervised learning in neural networks, beyond the shallow limit.

Dense Hebbian neural networks : a replica symmetric picture of unsupervised learning

Giannotti, Fosca;
2023

Abstract

We consider dense, associative neural-networks trained with no supervision and we investigate their computational capabilities analytically, via statistical-mechanics tools, and numerically, via Monte Carlo simulations. In particular, we obtain a phase diagram summarizing their performance as a function of the control parameters (e.g. quality and quantity of the training dataset, network storage, noise) that is valid in the limit of large network size and structureless datasets. Moreover, we establish a bridge between macroscopic observables standardly used in statistical mechanics and loss functions typically used in the machine learning. As technical remarks, from the analytical side, we extend Guerra's interpolation to tackle the non-Gaussian distributions involved in the post-synaptic potentials while, from the computational counterpart, we insert Plefka's approximation in the Monte Carlo scheme, to speed up the evaluation of the synaptic tensor, overall obtaining a novel and broad approach to investigate unsupervised learning in neural networks, beyond the shallow limit.
2023
Settore INF/01 - Informatica
Spin glasses; cost and loss functions; Hebbian learning
   Future Artificial Intelligence Research
   FAIR
   MUR
   PNRR

   Statistical mechanics of Learning Machines
   MUR
   PRIN2022

   Stochastic Methods for Complex Systems
   MUR
   PRIN2017
File in questo prodotto:
File Dimensione Formato  
Dense-Hebbian-neural-networks--A-replica-sy_2023_Physica-A--Statistical-Mech.pdf

Accesso chiuso

Tipologia: Published version
Licenza: Non pubblico
Dimensione 1.9 MB
Formato Adobe PDF
1.9 MB Adobe PDF   Richiedi una copia

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11384/137123
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 1
  • ???jsp.display-item.citation.isi??? 0
social impact