Valerio De Caro

Ph.D. Student in Computer Science

University of Pisa

Pervasive Artificial Intelligence Laboratory

I am a third-year Ph.D. student in Computer Science at the University of Pisa, under the supervision of Prof. Davide Bacciu and Dr. Claudio Gallicchio. My research focus is wrapped around the concept of learning in pervasive environments, merging Federated Learning and Continual Learning with methodologies for learning under resource constraints, including Reservoir Computing and Randomized Neural Systems.

Projects

FedRay

A Research Framework for Federated Learning based on Ray.

TorchESN

A PyTorch-based Implementation of Echo State Networks.

TEACHING

An EU-funded project with the aim of building a computing Toolkit for building Efficient Autonomous appliCations leveraging Humanistic INtelliGence. I am Task Leader for the human and system’s state monitoring task.

Featured Publications

Valerio De Caro, Claudio Gallicchio, Davide Bacciu

October, 2023 Neurocomputing

Continual Adaptation of Federated Reservoirs in Pervasive Environments

A method for adapting the internal dynamics of reservoirs in ESNs in a continual and federated environment.

Valerio De Caro, Antonio Di Mauro, Davide Bacciu, Claudio Gallicchio

October, 2023 In ESANN2023

Communication-Efficient Ridge Regression in Federated Echo State Networks

A communication-efficient approach to learning the readout in federated Echo State Networks (ESNs) by compressing the local information.

Valerio De Caro, Davide Bacciu, Claudio Gallicchio

July, 2023 ICML LLW 2023

Decentralized Plasticity in Reservoir Dynamical Networks for Pervasive Environments

A method for adapting the internal dynamics of reservoirs in ESNs in a continual and federated environment.

Vincenzo Lomonaco, Valerio De Caro, Claudio Gallicchio, Antonio Carta, Christos Sardianos, Iraklis Varlamis, Konstantinos Tserpes, Massimo Coppola, Mina Marmpena, Sevasti Politi, Erwin Schoitsch, Davide Bacciu

June, 2023 In ICASSP 2023 (Ambient AI Workshop)

AI-Toolkit: a Microservices Architecture for Low-Code Decentralized Machine Intelligence

An open-source microservices-based architecture for decentralised machine intelligence which aims at bringing R&D and deployment functionalities closer following a low-code approach

Valerio De Caro, Herbert Danzinger, Claudio Gallicchio, Clemens Koenczoel, Vincenzo Lomonaco, Mina Marmpena, Sevasti Politi, Omar Veledar, Davide Bacciu

June, 2023 In ICASSP 2023 (Ambient AI Workshop)

Prediction of Driver's Stress Affection in Autonomous Driving Simulations

An investigation to determine what is the best proxy of stress affection of a driver for performing prediction from physiological data.

See all publications

Contact

If you want to get in touch with me, just drop an e-mail!

valerio.decaro@phd.unipi.it
Largo Bruno Pontecorvo, 3, Pisa, PI 56127
Office 298a, Department of Computer Science, Floor 2