Allen Liu hopes to build a rigorous theoretical framework for understanding and developing machine learning algorithms.
A PhD student in computer science at the Massachusetts Institute of Technology (MIT), he is working toward developing algorithms with provable guarantees for a variety of fundamental learning problems in areas like robust statistics and preference learning. Having experienced firsthand the power and fragility of machine learning, Allen believes that a solid theoretical understanding is crucial to advancement of the field.
While robustness to noise has always been crucial in practical applications, it was not until recently that researchers have focused on developing algorithms with theoretical guarantees. In a recent breakthrough with Professor Ankur Moitra, Allen developed an efficient algorithm that provably learns the parameters of a mixture of Gaussians from samples, even in the presence of adversarial noise.
Allen received a bachelor’s degree in mathematics from MIT in 2020. Drawn to mathematics at a young age, he won numerous awards, including a perfect score at the International Mathematical Olympiad. A native of Rochester, New York, Allen enjoys playing soccer and is an expert alpine skier.
