Her research focuses on understanding learning and intelligence from first principles.
Tübingen – Siyuan Guo, a Ph.D. student in machine learning at the University of Cambridge and the Empirical Inference department at the Max Planck Institute for Intelligent Systems in Tübingen, received the award in recognition of her exceptional work in the field of machine learning and causality.
The MIT EECS Rising Star award recognizes Ph.D. students and postdocs in various domains of computer science. The Rising Stars in EECS, which was originally launched at MIT in 2012 and is being hosted by MIT and Boston University this year, is an annual award and workshop designed to help junior academic researchers advance their careers.
Siyuan Guo seeks to understand learning and intelligence from first principles. She finds intriguing partial answers at the intersection of physics and machine learning. Siyuan hypothesises that, similar to light travelling and Newtonian mechanics, learning also adheres to the laws of physics – specifically, the principle of least action. She derives learning algorithms such as the Bellman equation in reinforcement learning (RL) and Fisher-based optimisation (e.g. Adam) by seeking stationary trajectories in the Lagrangian. She is also working on the foundations of causality to understand invariant principles, and is developing a foundation model for causal inference in collaboration with the AI start-up Prior Labs.
Siyuan previously graduated from the University of Cambridge and University College London. Her research has been published at top-tier machine learning conferences (e.g. NeurIPS, ICLR), with one oral presentation and two spotlight presentations accepted. Due to positive participant feedback, she has been invited to lecture at the Cambridge Ellis Probabilistic Machine Learning Summer School in 2024 and again in 2025 to open the causality day. She has received funding and support from the G-Research PhD prize, the MPI-IS Outstanding Female Doctoral Student Award, the Cambridge-Tübingen fellowship and the Premium Research Studentship.