Good news on INNS BDDL 2019 program

The upcoming INNS Big Data and Deep Learning conference will be featuring two events from my group: a tutorial on Deep Learning for Graphs, jointly with Alessio Micheli, and an accepted oral paper survey deep learning models for tree transductions, jointly with my research associate Antonio Bruno.

Check out the INNS BDDL program here!

Bacciu Davide, Bruno Antonio: Deep Tree Transductions - A Short Survey. Proceedings of the 2019 INNS Big Data and Deep Learning (INNSBDDL 2019) , Forthcoming.

DeepDynamicHand paper now out

Check out my new joint work with Visar Arapi, Cosimo Della Santina, Matteo Bianchi and Antonio Bicchi. A deep learning based approach to recognize action primitives in video of object manipulation by humans. A first step towards automating robot manipulation skill acquisition from humans. 

Visar Arapi, Cosimo Della Santina, Davide Bacciu, Matteo Bianchi, Antonio Bicchi: DeepDynamicHand: A deep neural architecture for labeling hand manipulation strategies in video sources exploiting temporal information . In: Frontiers in Neurorobotics, 12 , pp. 86, 2018.

New graduates on adaptive tree processing

Congratulations (with some delay) to my students who graduated with two theses on the LISTIT project.

Valerio De Caro extended the Hidden Markov Tree Network and developed an efficient implementation of the model.

Michele Colombo developed a Conditional Variational Autoencoder for learning tree to tree  transductions with application to machine translation.

New Neurocomputing paper

Congratulations to my student Daniele Castellana for his first journal paper on learning infinite mixture of tree models for structured data clustering. The paper has been selected to be extended for journal publication from the ESANN 2018 best papers.

Bacciu Davide, Castellana Daniele: Bayesian Mixtures of Hidden Tree Markov Models for Structured Data Clustering. In: Neurocomputing (Accepted), Forthcoming.

Learning on Graphs and Explainable ML sessions @IJCNN 2019

I am co-organizing two special sessions at IJCNN 2019 next summer:

  • Special Session on Explainable Machine Learning, organized by Davide Bacciu (University of Pisa), Paulo J.G. Lisboa (Liverpool John Moores University), José D. Martín-Guerrero (Universitat de Valencia), Alfredo Vellido (Universitat Politècnica de Catalunya)
  • Special Session on Learning Representations for Structured Data, organized by Davide Bacciu (University of Pisa), Thomas Gärtner (University of Nottingham), Nicolò Navarin (University of Padova) and Alessandro Sperduti (University of Padova)

Paper submission deadline is 15 December 2018 (extension pending).

Send you work through the conference submission site!

New Ph.D. Student

Welcome Federico Errica, joining the Machine Learning group as a Ph.D. student under my joint supervision with Alessio Micheli.

Pleasure to be working with such a smart guy!

Tutorial on Deep Learning for Graphs at INNS BDDL 2019

Our tutorial on Deep Learning for Graphs will be presented at “INNS Big Data and Deep Learning 2019”: check out program and dates here.


Citizen Brain – AI&Politics Talk at Internet Festival

On Friday 12 October 2018 (15.00h) I will be giving a joint talk with my student Andrea Valenti at Internet Festival 2018 on the use of AI techniques to create fake political news and social posts (details follow)

The session will also host contributions from Lercio.

Event Details (in Italian)

Citizen Brain – La Comunicazione Politica al Tempo del Deep Learning

Venerdì, 12 ottobre 2018 –  15:00-18:30 – Centro Congressi Le Benedettine

L’oratoria politica è un’arte millenaria che crea consenso, movimenti di opinione e influenza le decisioni. Nel tempo si è evoluta nel genere, nei toni e nei canali di divulgazione. Oggi la comunicazione politica si trova spesso inserita quotidianamente nei social media, dove la produzione e pubblicazione di contenuti può essere automatizzata sfruttando tecnologie di Intelligenza Artificiale. Corriamo il rischio di campagne elettorali condotte da Bot politici? Sei (e forse più) discorsi politici in cerca d’autore ci sveleranno quanto siamo distanti dal realizzare un ghost writer artificiale.

L’evento è gratuito e liberamente accessibile sino a esaurimento posti. La prenotazione è obbligatoria per scuole e gruppi numerosi e consigliata per i visitatori singoli.

Special session on societal issues in machine learning @ESANN2019

I am co-organizing a special session at the upcoming ESANN 2019 on societal impacts of artificial intelligence, covering privacy, safety, ethical and fairness issues of machine learning.

We welcome contributions both from a theoretical and methodological side, as well as studies stemming from major research initiatives and projects focusing on the session topics.

Deadline for paper submission: 19 November 2018.

Prospective contributors/participants can contact me (or another co-organizer) for details.

Organized by Davide Bacciu (University of Pisa,Italy), Battista Biggio (University of Cagliari, Italy), Paulo J. G. Lisboa, (Liverpool John Moores University, U.K.), José D. Martín (Universitat de València, Spain), Luca Oneto (University of Genoa, Italy), Alfredo Vellido (Universitat Politècnica de Catalunya, Spain)

Paper on deep learning for tree transductions

A new paper for the LISTIT project has just been accepted for the IEEE SSCI 2018. Check out our sentence summarization approach using parse tree transductions!

Bacciu Davide, Bruno Antonio: Text Summarization as Tree Transduction by Top-Down TreeLSTM. Proceedings of the 2018 IEEE Symposium Series on Computational Intelligence (SSCI'18), IEEE, Forthcoming.