Please note: This PhD seminar will take place online.
Qianqiu Zhang, PhD candidate
David R. Cheriton School of Computer Science
Supervisor: Professor Ming Li
Artificial intelligence has generated enormous optimism in drug discovery, highlighted by AlphaFold’s Nobel Prize–winning success and bold predictions that AI could cure all diseases within a decade. Yet progress on individual computational tasks does not necessarily translate into therapeutic success. A persistent limitation is the field’s reliance on computational proxies that only partially capture the biological determinants of clinical efficacy and models can only optimize what has already been identified, formalized, and rendered measurable.
This seminar examines several open problems that illustrate this gap and raise the question of whether current AI paradigms can bridge local proxy-level optimization and integrated biological complexity.