Before joining the group, I graduated in physics in 2016 (BSc), with a thesis on using neural networks to spot dark photons at CERN, advised by Stefano Giagu;
and two years later in computer science (MSc), by defeating the Italian Othello champion with a cheap reimplementation of AlphaGo Zero from scratch, advised by Alessandro Panconesi.
Along with Roberto Di Leonardo, he was also my tutor at the formidable Sapienza School for Advanced Studies.
Before and during my PhD, I worked as a Research Intern for Spiketrap with Andrea Vattani, and as an Applied Scientist Intern for the Amazon Lablet in Tübingen with Francesco Locatello. Recently, I spent time with Alex Bronstein’s group at Technion in Haifa.
What makes us humans? I am interested in the intelligence that accounts for the difference existing between human beings and the other animals.
To me this gap is on par to the one between life and non-life and is the consequence of a fundamental property of nature, that should be understood in terms of information processing. Currently, I am persuaded that this intelligence coincides with our use of symbols, and that we shall model the mechanism through which humans link meaning to new signs. As when a scientist proposes a new theory.
Highlights of my research in this direction
- When a ML system becomes an artificial scientist: mastering the game of Zendo with Transformers.
Explanatory Learning: Beyond Empiricism in Neural Networks, 2022 [arXiv] [code] [Twitter thread] [Judea Pearl about this work].
- The meaning was already there: connecting text and images without training a neural network to do so.
ASIF: Coupled Data Turns Unimodal Models to Multimodal Without Training, 2022 [arXiv] [code] [Alex Smola about this work].
Other featured research
- It happens that different neural networks trained on the same stuff learn intrinsically equivalent latent spaces.
Relative Representations Enable Zero-shot Latent Space Communication, 2022 [arXiv] [code] [Oral at ICLR23 with 8-8-10 reviews].
- AlphaGo Zero for Othello. With two ideas to speed up the learning, and tested in a live match against a former world champion.
OLIVAW: Mastering Othello without Human Knowledge, nor a Penny, 2022 [arXiv] [Trailer of the match].
- With the right geometric prior, 11 samples are enough to train a generative model for 3D shapes of humans or animals.
LIMP: Learning Latent Shape Representations with Metric Preservation Priors, 2020 [arXiv] [code] [Oral at ECCV 2020 (2 min, 10min video)].
- The task with the widest gap between human and machine performance in BIG-bench, a collaborative effort to test Language Models.
Beyond the Imitation Game: Quantifying and Extrapolating the Capabilities of Language Models, 2022 [arXiv] [SIT task].
Check my Google Scholar profile for a complete list of my articles.
Selected Invited Talks
- 12/06/2023 Simons Institute at the University of California, Berkeley (US 🇺🇸). Opening talk at the workshop Decoding Communication in Nonhuman Species II, organized by David Gruber, Shafi Goldwasser, and Michael Bronstein. Explanatory Learning: Unleashing New Knowledge in Unknown Languages [video recording]
- 15/02/2023 Univeristy of Cambridge, CS department (UK 🇬🇧). ASIF: Coupled Data Turns Unimodal Models to Multimodal Without Training
- 12/09/2022 Tokyo Institute of Technology, CS department (Japan 🇯🇵). From sound to metric priors: new paradigms for shape generation
- 09/06/2022 Quantum Photonics, ONG (Virtual 🌐). How to create an artificial scientist
- 17/12/2021 International Symposium “Giornalismo e disinformazione”, UniPa (Palermo, 🇮🇹). Unire i puntini: meraviglie e limiti dei moderni sistemi di IA
- 12/06/2020 Cassini Junior Workshop, French Embassy in Italy and SSAS (Rome 🇮🇹). Towards a human-level artificial intelligence
- 19/02/2019 Italian Association for Machine Learning, ML meetup (Rome 🇮🇹): The italian AlphaZero [360° clip]
I enjoy mentoring younger students and coadvising them on their thesis with prof. Rodolà. If you are super passionate about AI and looking for what to do next or for a thesis (maybe an AI agent for a board game involving a cool challenge?) feel free to reach out at [my last name] at di.uniroma1.it!
Students I am advising/had advised on their BSc/MSc thesis:
- Robert Adrian Minut, MSc, Backward LLMs (Now PhD at Sapienza)
- Alessandro Zirilli, BSc, AlphaZero for Hex
- Ahmedeo Shokry, MSc, AI and Feynman Diagrams
- Giovanni Quadraroli, BSc, DeepRL for Space Invaders
- Guido Maria D’Amely Di Melendugno, MSc, DL for Contract Bridge (Now PhD at Sapienza)
- 22/09/2023 🎤 Invited talk at CENTAI (Turin, Italy 🇮🇹), hosted by the wonderful group of Giovanni Petri: Explanatory Learning: Towards Artificial Scientific Discovery.
- 21/09/2023 📰 Two papers accepted at NeurIPS 23: ASIF: Coupled Data Turns Unimodal Models to Multimodal without Training and Latent Space Translation via Semantic Alignment, see you in New Orleans! 🐊
- 08/08/2023 📏 UniReps Workshop accepted at NeurIPS 2023! We will talk about Unifying Representations in Neural Models in New Orleans this November. I will serve as panel chair.
- 28/07/2023 📰 Presented our poster at the KLR workshop at ICML (Honolulu, US 🇺🇸): Explanatory Learning: Towards Artificial Scientific Discovery
- 17/07/2023 🎤 Invited speaker at the workshop il Futuro Annunciato, to show the golden path of Computer Science to young students from physics and other hard sciences. (Rocca Sinibalda, Italy 🇮🇹)
- 14/07/2023 📍 Back in Rome from Haifa: end of my visit at Technion. Updates on my project there soon!
- 15/06/2023 💬 Invited research visit to the Stanford NLP group (Palo Alto, US 🇺🇸) hosted by Federico Bianchi to talk about ASIF.
- 14/06/2023 🎤 Invited talk at BAIR (US 🇺🇸), hosted by the Berkeley NLP Group: Explanatory Learning: Unleashing New Knowledge in Unknown Languages.
- 12/06/2023 🎤 Invited talk at the Simons Institute at the University of California, Berkeley (US 🇺🇸), for the workshop Decoding Communication in Nonhuman Species II, organized by David Gruber, Shafi Goldwasser, and Michael Bronstein. Explanatory Learning: Unleashing New Knowledge in Unknown Languages [video recording]
- 05/05/2023 🎤 Invited speaker at the panel Intelligenza artificiale: il racconto del cinema tra fantascienza e scenari of the 2023 Catania Book Festival (Catania, Italy 🇮🇹).
- 18/03/2023 📍 Moving to Haifa (Israel 🇮🇱) to start a research visit at Technion to work with Alex Bronstein.
- 21/02/2023 💬 Invited research visit to Deepmind (London, UK 🇬🇧) hosted by Marc’Aurelio Ranzato to talk about ASIF and RelReps.
- 20/02/2023 💬 Invited research visit to the University of Oxford (UK 🇬🇧) hosted by Michael Bronstein to talk about Explanatory Learning, ASIF, and RelReps [Emanuele giving the talk].
- 16/02/2023 🎤 Invited talk at Autodesk London (UK 🇬🇧) hosted by Hooman Shayani: ASIF: Coupled Data Turns Unimodal Models to Multimodal Without Training.
- 15/02/2023 🎤 Invited talk at the University of Cambridge (UK 🇬🇧) hosted by Pietro Liò: ASIF: Coupled Data Turns Unimodal Models to Multimodal Without Training.
- 14/02/2023 💬 Invited research visit to Imperial College London (UK 🇬🇧) hosted by Fabrizio Frasca to talk about ASIF and RelReps. [Luca’s talk before mine].
- 13/02/2023 💬 Invited research visit to University College London (UK 🇬🇧), hosted by Andrew Saxe to talk about RelReps. [Lunch selfie]
- 07/02/2023 💬 Invited research visit to BOSCH AI (Renningen, Germany 🇩🇪) to talk about applications of our latest work Relative Representations Enable Zero-shot Latent Space Communication (RelReps).
- 02-03/2023 👨🏫 Teaching assistant of the 2023 Deep Learning & applied AI course in Sapienza CS master degree. I was responsible for the first lab sessions (5 hours).
- 21/01/2023 📰 Relative Representations Enable Zero-shot Latent Space Communication accepted as notable-top-5% at ICLR-2023.
- 04/10/2022 📰 ASIF: Coupled Data Turns Unimodal Models to Multimodal Without Training is now on arXiv!
- 20/09/2022 🎤 Invited speaker at the symposium Saremo assimilati? Meraviglie, trappole e limiti dell’intelligenza artificiale, organized by the italian Order of Journalists (Cagliari, Italy 🇮🇹).
- 12/09/2022 🎤 Invited talk at the Tokyo Institute of Technology (Tokyo, Japan 🇯🇵) hosted by Asako Kanezaki: From sound to metric priors: new paradigms for shape generation
- 06/09/2022 📰 Errare humanum est? a pilot study to evaluate the human-likeness of a AI othello playing agent published at IVA 22
- 05/09/2022 📰 Presented our poster at the ml4evolang workshop at the JCoLE conference (Kanazawa, Japan 🇯🇵): Learning to make sense out of ambiguous messages leads to language evolution, a simulation
- 19/07/2022 🎤 Invited talk at DataScienceSeed (in italian, virtual): Explanatory Learning: può una macchina imparare a formulare teorie?
- 15/07/2022 📍 Back in Rome from Tübingen: end of my internship at Amazon (internship project: ASIF)
- 09/06/2022 ⭐ Our Symbol Interpretation Task in BIG-bench is the one with the widest gap between human and machine performance
- 09/06/2022 🎤 Invited talk at Quantum Photonics (virtual): How to create an artificial scientist
- 27/04/2022 👨🏫 Guest lecture for the Computer Science course in Sapienza Deep Learning and applied AI, titled Towards an artificial scientist.
- 08/03/2022 📰 OLIVAW: Mastering Othello without Human Knowledge, nor a Penny published on IEEE Transactions on Games
- 25/01/2022 📰 Explanatory Learning: Beyond Empiricism in Neural Networks is now on arXiv! Twitter thread
- 15/01/2022 📍 Moving to Tübingen (Germany 🇩🇪) to start a research internship at Amazon Science where I will work with Francesco Locatello.
- 17/12/2021 🎤 Invited speaker at the International Symposium Giornalismo e disinformazione at UniPa (Palermo, Italy 🇮🇹), panel Umano, troppo umano? Intelligenza Artificiale e disinformazione
- 08/11/2021 📏 Organizer of the Sapienza CS Doctoral Workshop 2021 in the fortress of the medieval village of Bertinoro (Italy 🇮🇹), a four-day summit of talks, activities, and networking.
- 02-05/2021 👨🏫 Teaching assistant of the 2021 Deep Learning & applied AI course in Sapienza CS master degree. I was responsible for the lab sessions (20 hours), this year we added a tutorial on Transformers, I curated the first part. I coauthored midterm and written exams also this year.
- 13/11/2020 🎤 Contributed talk at AAAI Fall Symposium on Abstraction and Analogy in AI (virtual): The value of a rationalist approach in AI
- 12/06/2020 🎤 Invited talk at Cassini Junior Workshop, organized by French Embassy in Italy and SSAS (Rome 🇮🇹): Towards a human-level artificial intelligence
- 27/03/2020 📰 LIMP: Learning latent shape representations with metric preservation priors accepted as oral at ECCV-2020.
- 26/02/2020 🎤 Contributed talk at Technion - Israel Institute of Technology (Haifa, Israel 🇮🇱) Learning deformable style transfer via differentiable intrinsic distances
- 02-05/2020 👨🏫 Teaching assistant of the 2020 Deep Learning & applied AI course in Sapienza CS master degree. I was responsible for the lab sessions (20 hours), coauthoring 10 tutorials and the written exams.
- 01/11/2019 📍 Started my PhD at Sapienza University of Rome advised by prof. Emanuele Rodolà
- 19/02/2019 🎤 Invited talk at the ML meetup from the Italian Association for Machine Learning (Rome 🇮🇹): The italian AlphaZero