Hello. I’m Ilia.

I’m a Faculty Fellow/Assistant Professor at the NYU Center for Data Science.

Before that, I was a postdoc in the CoCoSci Lab at Princeton University supervised by Dr. Tom Griffiths.

And before that, I did my PhD in Statistics at the University of Waterloo where I defended my thesis on ‘Learning From Almost No Data’ supervised by Dr. Matthias Schonlau.

I use tools like representational alignment to study the parallels between human and artificial intelligence in order to probe the limits of learning/teaching with small data and develop AI thought partners that learn & think with people rather than instead of people.

Check out my publications to see my progress so far. If you find something you’re interested in discussing then shoot me an email and I’d be happy to chat. The best place to reach me is at <is3060 [at] nyu [dot] edu>.

News

December 2024: We’re organizing the BehavioralML Workshop at NeurIPS 2024!

September 2024: I’m starting as a Faculty Fellow/Assistant Professor at NYU CDS!

June 2024: We’re organizing the LLMs and Cognition Workshop at ICML 2024!

May 2024: We’re organizing the AI session at Neuromonster 2024!

May 2024: We’re organizing the Re-Align Workshop at ICLR 2024!

December 2023: Our paper, “Alignment with human representations supports robust few-shot learning”, was accepted as a spotlight at NeurIPS 2023!

August 2023: Our papers, “Human-in-the-loop mixup” and “On the informativeness of supervision signals”, were accepted, respectively, as an oral and spotlight at UAI 2023!

July 2023: We’re organizing the LLMs Meet Cognitive Science Workshop at CogSci 2023!

March 2022: I received an NSERC Postdoctoral Fellowship to support my research on STEM-AI: Scientific Theory Encoding Methods for Artificial Intelligence.

January 2022: I’m also joining Josh Tenenbaum’s CoCoSci Lab at MIT as a research affiliate.

July 2021: I’m joining Tom Griffiths’s CoCoSci Lab at Princeton University as a postdoc.

June 2021: I defended my thesis on ‘Learning From Almost No Data’.

May 2021: I’m joining Kin-Keepers as an AI Advisor to support their mission of developing a solution that will enable dementia sufferers to communicate and feel understood.

Apr 2021: Two of our papers were accepted to IJCNN 2021: “Soft-Label Dataset Distillation and Text Dataset Distillation” (preprint) and “One Line To Rule Them All: Generating LO-Shot Soft-Label Prototypes” (preprint)

Jan 2021: Our reseach was profiled on Scientific American as a pathway towards democratizing AI.

Oct 2020: Our LO-shot learning research was featured on MIT Tech Review, Digital Trends, and several other outlets!

Aug 2020: I’m joining Stratum AI as VP of Research and will be leading the development of ML/DL methods to make mining more efficient and environmentally sustainable.