Paolo Frasconi

I'm a full professor of Computer Science at DINFO, University of Florence. I have been an Associate Professor at Dipartimento di Sistemi e Informatica from 2000 to 2010, and an Associate Professor at the University of Cagliari from 1998 to 2000. I've been a Visiting Professor at KU Leuven in 2010, a Lecturer at the University of Wollongong, Australia in 1998, and a Visiting Scholar at MIT in 1992.

My research interests are in the area of machine learning, with particular emphasis on algorithms for structured and relational data, learning with logical representations, and applications to bioinformatics. In these areas I've written over 180 refereed papers. You may find my list of publications on my Scholar profile and on this page.

Among other activities, I'm an Associate Editor of the Machine Learning Journal. I co-chaired the European Conference on Machine Learning and Principles and Practice of Knowledge Discovery (2016), the Int. Conference on Prestigious Applications of Artificial Intelligence (PAIS) 2012, the AAAI 2010 Special Track on AI and Bioinformatics, the 20th International Conference on Inductive Logic Programming (2010), and the 5th International Workshop on Mining and Learning with Graphs (2007).

I'm head of the regional PhD Program in Smart Computing of the Universities of Florence, Pisa and Siena.


I'm interested in machine learning and its applications. In particular I've been working on recurrent and recursive neural networks, kernel methods, and graphical models for learning with structured/relational data. More recently I have been working on hyperparameter optimization and meta-learning. I'm also interested in bridging statistical learning and symbolic (logic-based) approaches. These ideas have led to systems such as kFOIL (Landwher et al. 2006; 2010), type extension trees (Frasconi et al. 2008; Jaeger et al. in preparation), and more recently kLog (Frasconi et al. 2012). Application areas of my interest include bioinformatics (in particular protein structure and function, molecular activity), natural language processing, computer vision.

Servers & software

is a TensorFlow framework for gradient-based hyperparameter optimization and meta-learning developed by Luca Franceschi and Riccardo Grazzi (see also the AutoML 2018 paper)
Brain cell finder
is a tool for fully automated localization of soma in 3D mouse brain images acquired by confocal light sheet microscopy
is a logical and relational language for kernel-based learning embedded in Prolog.
converts bitmap images of chemical structural formulae into machine readable vector formats (such as MOL and SDF).
A predictor of protein metal binding state. Source code, Dataset
A predictor of cysteines bonding state and disulfide bridges.
A standalone version of DISULFIND, available on Ubuntu 12.04+ and Debian Sid+
Learning simple relational kernels (written by Niels Landwehr and Andrea Passerini)
Type Extension Trees
a powerful representation language for "count-of-count" features characterizing the combinatorial structure of neighborhoods of entities in relational domains (written by M. Lippi).
Markov Logic Networks with Grounding-Specific Weights, a mod of the Alchemy system (written by M. Lippi).
3D and 2D Decomposition Kernels for classification of small molecules (written by A. Ceroni and F. Costa).
Weighted Decomposition Kernels (C++ Implementation; written by F. Costa).

Some talks

ACAI 2018
Kernels and Deep Networks for Structured Data. Lecture at Summer School on Statistical Relational Artificial Intelligence 2018, Ferrara
DeepLearn 2018
Bilevel Programming for Hyperparameter Optimization and Meta-Learning. Invited keynote at DeepLearn 2018, Genova
Gradient-based Hyperparameter Optimization. Invited talk at 11th International Conference on Energy Minimization Methods in Computer Vision and Pattern Recognition 2017, Venice
Shonan 2014
Gradient-based Hyperparameter Optimization. Shonan Meeting 040 - Deep Learning: Theory, Algorithms, and Applications, 2014, Shonan, Japan
CoLISD 2011
kLog: A Language for Logical and Relational Learning with Kernels. Invited talk at CoLISD 2011 (ECML/PKDD Workshop on Collective Learning and Inference on Structured Data), Athens
Dagstuhl 2009
Learning protein metal binding Dagstuhl Seminar on Similarity-based learning on structures, Feb. 2009
ILP 2007
Learning with Kernels and Logical Representations Invited keynote at the 17th Int. Conference on Inductive Logic Programming (ILP'07)



Frasconi, Paolo, Niels Landwehr, Giuseppe Manco, and Jilles Vreeken. Machine Learning and Knowledge Discovery in Databases. European Conference, ECML PKDD 2016, Riva del Garda. LNAI 9851-9852. Springer. ISBN 978-3-319-46128-1

ECAI 2012

De Raedt, Luc, Christian Bessiere, Didier Dubois, Patrick Doherty, Paolo Frasconi, Fredrik Heintz, and Peter J. F. Lucas, (2012). ECAI 2012 - 20th European Conference on Artificial Intelligence. Including Prestigious Applications of Artificial Intelligence (PAIS-2012) System Demonstrations Track, Montpellier, France, August 27-31 , 2012. ECAI Frontiers in Artificial Intelligence and Applications. IOS Press

ILP 2010

Frasconi, Paolo and Francesca A. Lisi (2011). Inductive Logic Programming: 20th International Conference, ILP 2010, Florence, Italy, June 27-30, 2010, Revised Papers. Springer. ISBN 978-3-642-21294-9

Probabilistic ILP

De Raedt, Luc, Paolo Frasconi, Kristian Kersting, and Stephen Muggleton (2008). Probabilistic Inductive Logic Programming --- Theory and Applications. Springer. ISBN 978-3-540-78651-1

Modeling the Internet

Baldi, Pierre, Paolo Frasconi, and Padhraic Smyth (2003). Modeling the Internet and the Web: Probabilistic Methods and Algorithms. John Wiley \& Sons. ISBN 978-0-470-84906-4


Frasconi, Paolo and Ron Shamir (2003). Artificial Intelligence and Heuristic Methods for Bioinformatics. IOS Press. ISBN 978-1-58603-294-4




Area chair

PC member/reviewer (recent)

Other past events


Services for my university


Office hours

Tuesday, 10:45-12:45

Current courses

Fall 2019 B024317 - Machine Learning
Laurea Magistrale in Ingegneria Informatica (B070), Laurea Magistrale in Informatica (B059)
Fall 2019 B003725 - Intelligenza Artificiale (Artificial Intelligence)

Past courses

Fall 2018 B024317 - Machine Learning
Laurea Magistrale in Ingegneria Informatica (B070), Laurea Magistrale in Informatica (B059)
Fall 2018 B003725 - Intelligenza Artificiale (Artificial Intelligence)
Laurea in Ingegneria Informatica (B047)
Spring 2018 Hyperparameter optimization, meta-learning, and all that
PhD Program in Smart Computing
Fall 2017 B024317 - Machine Learning
Laurea Magistrale in Ingegneria Informatica (B070), Laurea Magistrale in Informatica (B059)
Fall 2017 B003725 - Intelligenza Artificiale (Artificial Intelligence)
Laurea in Ingegneria Informatica (B047)
Fall 2016 B024317 - Machine Learning
Laurea Magistrale in Ingegneria Informatica (B070), Laurea Magistrale in Informatica (B059)
Fall 2016 B003725 - Intelligenza Artificiale (Artificial Intelligence)
Laurea in Ingegneria Informatica (B047)
July 2016 Machine Learning
Master on Big Data Analytics and Technologies for Management
April 2016 Deep Learning
PhD Program in Computer Science and Computational Mathematics. Univ. Insubria
March 2016 Deep Learning
Doctoral School in Information and Communication Technology, Univ. Trento
February 2016 Deep Learning
Doctoral Program in Smart Computing, Univ. Firenze


DINFO — Università degli Studi di Firenze.
Via di Santa Marta 3, I-50139 Firenze, Italy.

+39 (055) 275-8647

Directions. Public transport: From Firenze main train station (Santa Maria Novella) take tram line 1 towards Careggi, get off at Poggetto and then take bus 55 at Celso (3rd stop).

Once there... The Santa Marta building is beatiful but rather old. Room numbers do not follow any rational pattern and finding your way may be a bit challenging. The plan below may help (note however it was done before renewing). Both my office (red) and my lab (green) are shown.