7 papers
Exponential Capacity in Multilayer Hetero-Associative Neural Networks
Elena Agliari, Adriano Barra, Andrea Ladiana +1
Exponential Hopfield networks store a number of patterns that grows exponentially with the number of neurons, and in their classical formulation they are auto-associative: they com…
Semi-supervised Hopfield model: Theoretical and Numerical results
Linda Albanese, Andrea Ladiana, Andrea Lepre
In the daily practice of Machine Learning, fully labeled datasets are a luxury: labels demand expensive and time-consuming human annotation, whereas raw, unlabeled data can be harv…
Finite-size scaling of hetero-associative retrieval in continuous-signal-driven Ising spin systems
Andrea Ladiana
Real-world physical signals are continuous and high-dimensional, yet the statistical-mechanics machinery of associative memory operates on discrete Ising spins. We bridge this divi…
Do Hopfield Networks Dream of Stored Patterns? A Statistical-Mechanical Theory of Dreaming in Multidirectional Associative Memories
Adriano Barra, Fabrizio Durante, Andrea Ladiana +1
We introduce the Dreaming -directional Associative Memory (DLAM), a multi-layer Hebbian architecture in which off-line dreaming and supervised heteroassociative coupling coexist…
A Federated Many-to-One Hopfield model for associative Neural Networks
Andrea Alessandrelli, Fabrizio Durante, Andrea Ladiana +1
Federated learning enables collaborative training without sharing raw data, but struggles under client heterogeneity and streaming distribution shifts, where drift and novel data c…
Supervised and Unsupervised protocols for hetero-associative neural networks
Andrea Alessandrelli, Adriano Barra, Andrea Ladiana +2
This paper introduces a learning framework for Three-Directional Associative Memory (TAM) models, extending the classical Hebbian paradigm to both supervised and unsupervised proto…