Scan contact card

Scannable QR code with contact info

PhD Student at Edinburgh Centre for Robotics busy trying to teach machines how to learn language through natural language interaction in multi-modal environments. I am mainly interested in developing Machine Learning models able to learn conceptual representations with few training examples by exploiting multiple modalities such as vision, language and proprioceptive information.

Skills

Machine Learning & Data Mining
  • Weka
  • TensorFlow
  • PyTorch
  • AllenNLP
  • SPaCy
  • RASA-NLU
  • scikit-learn
Languages & Tools
  • Java
  • Python
  • git
DBMS
  • MySQL
  • PostgreSQL
  • Amazon DynamoDB
  • MongoDB
  • Amazon Neptune
  • Neo4j
Semantic Web Technologies
  • SPARQL
  • OWL

Projects

A collection of interesting projects that I have been involved with during my academic career or that I have completed during my free time.

Alexa Prize Challenge 2018

Main member of the team 'Alana' responsible of NLU pipeline and several content-based bots.

Read more...

Ask Me Any Rating: A Content-based Recommender System based on Recurrent Neural Networks

Code for the paper 'Ask Me Any Rating: A Content-based Recommender System based on Recurrent Neural Networks'

Read more...

Contribution to the AllenNLP project

Fixed bug related to the attention mechanism implemented in AllenNLP

Read more...

Integrated Dialog bAbI+ dataset in ParlAI

Integrated Dialog bAbI+ dataset in the Facebook ParlAI framework

Read more...

Iterative Multi-document Neural Attention for Multiple Answer Prediction

Code for the paper 'Iterative Multi-document Neural Attention for Multiple Answer Prediction'

Read more...

Schema-aware Translating Embeddings model

Code for the university project in 'Semantic Web Technologies'

Read more...

Publications

A list of recent publications made by myself and my collaborators. Please refer to my Google Scholar for an updated and complete list.

Paper - Entertaining and Informative Open-domain Social Dialogue using Ontologies and Entity Linking

Conversational AI system classified 3rd place in the Amazon Alexa Prize Challenge 2018

November 2018

Paper - An automatic procedure for generating datasets for conversational recommender systems

Proceedings of Dynamic Search for Complex Tasks-8th International Conference of the CLEF Association, CLEF

November 2017
July 2017

Paper - Iterative multi-document neural attention for multiple answer prediction

Deep Understanding and Reasoning: A challenge for Next-generation Intelligent Agents (URANIA) workshop, held in the context of the AI*IA 2016 conference

February 2017

Experience

PhD Student

Edinburgh Centre for Robotics

Research project focused on Interactive Grounded Language Learning

September 2017 - Present

ML/NLP Consultant

Plusimple S.r.l

Developed search engine and analytics platform for the Plusimple userbase.

March 2017 - September 2017

Education

University of Bari Aldo Moro

Bachelor of Computer Science (Full marks with Honors)
Computer Science
Thesis title: 'Content-based Recommender Systems based on Linked Open Data'
2011 - 2014

University of Bari Aldo Moro

Master of Computer Science (Full marks with Honors)
Knowledge Engineering and Machine Intelligence
Thesis title: 'Converse-Et-Impera: a Conversational Content-based Recommender System based on Hierarchical Deep Reinforcement Learning'
2014 - 2017

Heriot-Watt University & University of Edinburgh

Master by Research (Full marks with Distinction)
Robotics and Autonomous Systems
2017 - 2018
Nifty tech tag lists from Wouter Beeftink