| KEYWORD |
Design and implementation of a software platform to manage Artificial Neural Networks
Parole chiave ALGORITMI, ANN, MACHINE LEARNING, NEURONAL NETWORK
Riferimenti ANDREA ACQUAVIVA, EDOARDO PATTI
Riferimenti esterni Alessandro Aliberti (alessandro.aliberti@polito.it)
Gruppi di ricerca ELECTRONIC DESIGN AUTOMATION - EDA, GR-06 - ELECTRONIC DESIGN AUTOMATION - EDA, ICT4SS - ICT FOR SMART SOCIETIES
Tipo tesi SPERIMENTALE
Descrizione The aim of the project is designing and building a toolbox manager for Artificial Neural Network (ANN). Starting from methodologies and algorithms already implemented in Matlab, the student will: i) perform a comparative analysis of the current machine learning platforms (e.g. Tensorflow, PyTorch, etc.) identifying which platform is more versatile and performant; ii) implement the methodologies and algorithms on the new platform. This process will create a new platform for the management of neural networks to model and control dynamic systems.
Scadenza validita proposta 30/03/2020
PROPONI LA TUA CANDIDATURA