Research Lead, Iliad

Zach Furman is a research lead at Iliad, working on singular learning theory and the mathematical foundations of deep learning. He is also a PhD candidate at the University of Melbourne, supervised by Liam Hodgkinson and Daniel Murfet. He has a bachelor's degree in mathematics and computer science from Boston University, and prior experience in aerospace engineering and condensed matter physics.
Projects surround theoretical understanding of deep learning, in particular training dynamics and singular learning dynamics. For examples of prior research in this category:
- https://arxiv.org/abs/2506.06489
- https://arxiv.org/abs/2411.19920
- https://arxiv.org/abs/2302.11055
- https://arxiv.org/abs/2308.12108
- Interest in scientific/theoretical understanding of deep learning
- Mathematical maturity: the ability to understand and develop mathematical concepts, at the level of a math/physics Master's degree or equivalent experience
- Research maturity: the ability to independently identify promising hypotheses, refine them with theoretical arguments, and test via experiments
- High levels of autonomy and independence
- Experience with algebraic geometry is very helpful for some projects