Dr. Eric Anschuetz

Burke Fellow at Caltech

Speaker: Dr. Eric Anschuetz Date: 22-01-2026, 11am-12pm (BST) Location: Department of Computer Science CS1.04, University of Warwick, Coventry, UK

What Goes Wrong in Quantum Machine Learning (and How to Make it Right)

Abstract

Quantum computers use the principles of quantum mechanics to perform computations in a way fundamentally different from traditional computational methods. This new computing paradigm exhibits great promise in–among other things–performing linear algebra and sampling extremely efficiently, naturally making the development of new machine learning models with an eye toward quantum implementations an exciting prospect. In this talk, I will lay out the important differences between quantum and traditional machine learning models, share why much of our intuition from machine learning fails for these new models, and show that structure is a necessary ingredient for building efficiently-trainable quantum neural networks.

---

About Dr. Eric Anschuetz

Eric is a Burke Fellow at Caltech who completed his PhD at MIT under the joint supervision of Aram Harrow and Misha Lukin. Much of his research involves studying the limitations of quantum algorithms through the lens of statistical physics and quantum foundations theory. These insights have led to the development of novel quantum algorithms for optimization and learning, and a new, physically-motivated approach to quantum computational complexity.