1.

Record Nr.

UNINA9911034857903321

Autore

Stipanović Dusan

Titolo

Difference Equations and Machine Learning / / by Dušan Stipanović

Pubbl/distr/stampa

Cham : , : Springer Nature Switzerland : , : Imprint : Springer, , 2025

ISBN

3-032-00910-3

Edizione

[1st ed. 2025.]

Descrizione fisica

1 online resource (187 pages)

Collana

Synthesis Lectures on Mathematics & Statistics, , 1938-1751

Disciplina

515.625

515.75

Soggetti

Difference equations

Functional equations

Machine learning

Artificial intelligence

Mathematical analysis

Neural networks (Computer science)

Mathematics

Difference and Functional Equations

Machine Learning

Artificial Intelligence

Analysis

Mathematical Models of Cognitive Processes and Neural Networks

Applications of Mathematics

Lingua di pubblicazione

Inglese

Formato

Materiale a stampa

Livello bibliografico

Monografia

Nota di contenuto

Introduction -- Linear Difference Equations -- Nonlinear Difference Equations -- Stability and Chaotic Behaviors of Difference Equations -- Control of Difference Equations -- Applications to Neural Networks and Machine Learning -- Conclusions.

Sommario/riassunto

This book presents in-depth explanations of well-known and recognized behaviors of neural networks in machine learning. In addition, the author provides novel technical analyses of behaviors of discrete-time dynamical systems modeled as difference equations. These analyses and their outcomes are closely related to models of very



well-known neural networks such as Long Short-Term Memory (LSTM) and Gated Recurrent Unit (GRU) neural networks, which are widely used in machine learning and artificial intelligence (AI) applications. The author also discusses difference equations and their relevance to neural networks, machine learning, and AI. In addition, this book: Includes characterizations of difference equations and technical prospectives of discrete-time systems Provides new insights into the dynamical behaviors of some of the most popular neural networks used in machine learning Discusses novel technical analyses of discrete-time dynamical systems modeled as difference equations.