I am a machine learning researcher, curious about the nature of intelligence. I’ve been fortunate to spend time at the University of Cambridge with Ferenc Huszár, the Max Planck Institute for Intelligent Systems with Bernhard Schölkopf, and interned at Meta FAIR in New York.

My long-term interest is AI + Science: helping build a physics-inspired mathematical theory of intelligence and using AI to accelerate scientific discovery. My work is driven by two intertwined questions:

  1. How can principles from nature guide the design of learning systems?
    This underpins my recent physics of learning programme, illustrating that learning too follows the least action principle and classic learning algorithms are derivable.

  2. How can we identify fundamental laws of nature directly from data?
    This includes causal de Finetti line of work on the Bayesian foundations of causality and building a foundation model for in-context causal inference on scientific applications.

Prospective students. If you are interested in working with me on the science of AI and/or AI for Science, please reach out with your CV and a short research statement.

Email: sguo26v at gmail dot com

Students and Alumni:

– Anna Kerekes (Co-mentored with B. Schölkopf)– Yuche Gao (Co-advised with J. M. Hernández-Lobato)
– Michelle Chao Chen (Co-advised with B. Schölkopf)
Structural Information in LLMs. Under Submission, 2026		– Bálint Mucsányi (Co-mentored with P. Reizinger)
Identifiable Representations for RL. ICLR, 2026
– Moritz Miller (Co-advised with J. Peters and B. Schölkopf)
In-context Counterfactual Reasoning. NeurIPS, 2025		– Szilvia Ujváry (Co-mentored with F. Huszár)

News

Interests
  • Physics of Learning
  • Machine Learning
  • AI for Science
  • Causality
Education

  • PhD in Machine Learning, 2021

    University of Cambridge and Max Planck Institute for Intelligent Systems

  • Msc in Machine Learning, 2020

    University College London

  • Bachelor and Master in Mathematics, 2015

    University of Cambridge