I find it interesting to replicate our skills with machines to realize what does (not) really make us humans.

Bio. Currently, CS postdoc at the University of Oxford trying to understand what the whales are saying through machine learning with Michael Bronstein.

I received my PhD degree in Computer Science cum laude from the Sapienza University of Rome, with a thesis on “Artificial Scientific Discovery” (video). In Rome, I worked in the GLADIA research group, advised by Emanuele Rodolà. During my PhD, I also worked as an Applied Scientist Intern for the Amazon Lablet in Tübingen with Francesco Locatello, and I spent time with Alex Bronstein’s group at Technion in Haifa. I was also an ELLIS student.

Before starting my PhD, I graduated in Physics in 2016 (BSc), with a thesis about 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 pivotal Sapienza School for Advanced Studies. In this period, I also worked as a Research Intern for Spiketrap with Andrea Vattani.

What makes us humans?

I am interested in understanding what allowed humans to reach the Moon. What do we have that animals and machines are missing?

To me this gap is the consequence of a fundamental property of nature, which we can understand by figuring out how to build machines that match our skills. At long last, we have machines capable enough, and we can submit our models of human skills to experimental evaluation. This is a groundbreaking moment in science: AI is doing now what physics did centuries ago, gloriously taking from philosophy the duty of explaining some fundamental aspects of reality, such as what knowledge is and how it is created.

Currently, I am obsessed with our use of symbols, especially by the mechanism through which humans link meaning to new signs, such as when a redditor creates a new meme, or a scientist proposes a new theory.

I want to build machines capable of writing scientific articles about their new ideas, pushing our knowledge forward.

Highlights of my research in this direction

  • A small model of 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][Poster ↓].
  • 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][Poster ↓].

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

My posters

Posters of my highlighted research papers, as they appeared in ICML 2023 and NeurIPS 2023. Click to see them in their full glory!

EL poster ASIF poster

Advising

I enjoy mentoring younger students and coadvising them on their thesis. 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 (Now PhD at École Polytechnique)
  • Giovanni Quadraroli, BSc, DeepRL for Space Invaders
  • Guido Maria D’Amely Di Melendugno, MSc, DL for Contract Bridge (Now PhD at Sapienza)

News