Stefano Melacci

Short Bio

Stefano Melacci is an Associate Professor of the Department of Information Engineering and Mathematics (DIISM) of the University of Siena (Siena, Italy), since January 2021. His research activity is focussed on Machine Learning and, more generally, Artificial Intelligence. He has over 15 years of experience with Machine Learning, covering both foundational aspects of learning and applications (mostly Language and Vision).

  • From 2018 to 2020 he was a Senior Researcher (RTDB) at DIISM, University of Siena.
  • In 2017 he received the national qualification (Abilitazione Scientifica Nazionale ASN) for becoming Associate Professor (09/H1 – ING-INF/05).
  • From 2015 to 2017 he has been working as researcher in the industry, mostly at QuestIT (Siena, Italy), dealing with Natural Language Processing problems. In particular, in 2015-2016 he has been involved in the study and definition of innovative Machine Learning-based technologies for Conversational Systems. During 2017 he managed the research activity of the company (Research Manager), enriching the technology stack with (multi-language) Machine Learning-based solutions and improving the already existing Semantic Analysis algorithms.
  • From 2010 to 2014 he has been a Research Associate (Post-Doc) of the Department of Information Engineering (DII) / Department of Information Engineering and Mathematics (DIISM), University of Siena (Siena, Italy), under the supervision of Prof. Marco Gori (4 years) and Prof. Alessandro Agnetis (1 year). His research activity covered Machine Learning algorithms from a foundational perspective. He proposed multi-layer architectures (Deep Networks) for the extraction of information from static images and videos, through the use of adaptive convolutional filters and principles of Information Theory (Developmental Visual Agents (DVA)). He has been able to apply these technologies in the project “Learning to see like children”, with the aim of implementing a human-machine interaction protocol for the progressive development of visual skills. Many of his scientific proposals are about Kernel Machines and Regularization Theory (classification problems with supervised intervals, clustering and verification problems using minimal entropy criteria, …), described by means of the generic framework of “Learning from Constraints”, which deals with the integration of Machine Learning techniques (Kernel Machines, Deep Networks, …) and Symbolic Knowledge Representation & Reasoning.
  • From 2006 to 2009 he followed a Ph.D. program in Machine Learning (Adaptive Systems for Information Processing) at the Department of Information Engineering (DII) of the University of Siena (Siena, Italy), under the supervision of Prof. Marco Maggini. He received his Ph.D. degree in April 9, 2010. In 2009 he has been a Visiting Scientist (6 months) at the Computer Science and Engineering Department (CSE) of the Ohio State University (Columbus, Ohio, USA), under the supervision of Prof. Mikhail Belkin. He has been working on Manifold Regularization algorithms and proposed Neural Networks for tackling Similarly Measure Learning problems, with application to Computer Vision.
  • In 2006 he received the Master Degree in Computer Science (Ingegneria Informatica) from the University of Siena (Siena, Italy), with a thesis on the automatic generation of caricatures of human pictures.


Google Scholar Profile

Personal Home Page

Here is a link to my personal website


“Another visitor. Stay awhile… staaaaay FOREVER!” 🙂

Updated: December, 2020