Work

Constraint satisfaction approach in structuring neural network architectures

Year: 2025

Type: article

Abstract: This work presents a novel numerical and quantitative methodology grounded in Constraint Satisfaction Problem (CSP) theory, aimed at developing a specialized tool for the structural analysis of fully connected, feed-forward Neural Networks (NNs). The proposed approach enables a systematic exploration of neuron configurations within the hidden layers. A backtracking search algorithm was specifically designed to traverse the space of admissible architectural parameters, thereby implementing a constrained combinatorial strategy for neural network architecture exploration. This study introduces a practical tool for researchers aiming to identify diverse neuronal organizational patterns within hidden layers, subject to predefined hyperparameter constraints. The proposed algorithm was subsequently validated by exhaustively exploring all feasible architectural configurations for solving a two-dimensional Poisson equation using a Physics-Informed Neural Network (PINN).

Source: Journal of Computational and Applied Mathematics

DOI: 10.1016/j.cam.2025.117140

Authors: Vittorio Bauduin, Salvatore Cuomo, Vincenzo Schiano Di Cola

Institution: University of Naples Federico II

Language: English

Topic: Neural Networks and Applications

Subfield: Artificial Intelligence

Field: Computer Science

Domain: Physical Sciences

Open Access status: hybrid

Estimated APC paid: $3,250

Funders: European Commission, Ministero dell’Istruzione, dell’Università e della Ricerca, Ministero dell'Università e della Ricerca, Gruppo Nazionale per il Calcolo Scientifico