BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Match\, Train\, Improve: From Matched Data to Local Policy Improve
 ment in Molecular Design - Natasa Tagasovska (Roche Pharmaceuticals)
DTSTART:20260319T153000Z
DTEND:20260319T163000Z
UID:TALK244483@talks.cam.ac.uk
DESCRIPTION:Scientific discovery often operates in a regime where evaluati
 ons are expensive\, data are limited\, and useful interventions must remai
 n close to known good candidates. In this talk\, I present a matched-data 
 approach to molecular property enhancement that is naturally connected to 
 reinforcement learning. The key idea is to construct local pairs in which 
 one molecule is both nearby and better than another\, and to train a model
  to learn these local improving moves. Iterating this operator yields a pr
 actical strategy for lead optimization.I will argue that this method is be
 st viewed as a critic-free\, offline\, local policy-improvement procedure 
 rather than full RL. This framing clarifies its relationship to supervised
  fine-tuning and direct preference optimization: all three methods learn f
 rom paired data\, but matched training uses locality as an additional indu
 ctive bias\, allowing each pair to convey directional information about ho
 w to improve. I will also discuss extensions based on generative modeling 
 over matched datasets\, self-training\, and robust out-of-distribution gen
 eralization\, and conclude with opportunities for active data collection a
 nd uncertainty-aware planning in scientific design.
LOCATION:Seminar Room 1\, Newton Institute
END:VEVENT
END:VCALENDAR