BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Memento EPFL//
BEGIN:VEVENT
SUMMARY:Solving Scale Up in the Chemical Industries: Identifying the Bigge
 st Challenges and the Impact of Machine Learning
DTSTART:20260217T170000
DTEND:20260217T180000
DTSTAMP:20260530T044103Z
UID:f5b6cf2133414ad39044d79637b19eb0597eb3df80e2a1c8a247aa91
CATEGORIES:Conferences - Seminars
DESCRIPTION:Dr. José Folch (CSO\, SOLVE Chemistry)\, Dr. Linden Schrecker
  (CEO\, SOLVE Chemistry)\nMachine learning has changed the paradigm of lab
 oratory chemistry\, with a large drive into using algorithms to make decis
 ions\, optimize\, and analyse experimental results. The largest impact of 
 this has been made for the discovery of new materials\, drugs\, and agroch
 emicals with billion-valuation companies emerging to dominate the scene su
 ch as Isomorphic Labs and Cusp AI. However\, the scale up and process deve
 lopment of new products has been left behind by AI for Science. At SOLVE C
 hemistry\, we are looking to bring the benefits of machine learning to sca
 le-up science. Driven by the idea that true impact is blocked by a lack of
  data\, we have built an automated laboratory powered by a stack of novel 
 machine learning algorithms with the long-term goal of first-time right sc
 ale up\, bringing more new chemicals to market\, more sustainably\, and le
 ss expensive.\n\nWe will first briefly explain transient flow\, an emergin
 g methodology that allows us to efficiently gather data by slowly varying 
 conditions of a reactor and constantly measuring the stream of data. While
  the technique allows us to be very efficient at collecting data\, it pose
 s new challenges for Bayesian optimisation and active learning in such rea
 ctors\, which necessitated the development of transition-constrained metho
 ds that can plan ahead whole reactor campaigns\, acquiring the best data i
 n an order that minimizes the amount of steady-state waiting times. We wil
 l showcase the algorithm in practice\, used in a solvent replacement campa
 ign\, and explain further developments and challenges that arise. We will 
 then look at how Large Language Models can be used to generate insights af
 ter the reaction data is collected\, creating not an optimization loop\, b
 ut a scientific discovery loop where the target becomes the discovery of t
 rue underlying chemistry that is scale-independent and can be used to spee
 d up process development.\n 
LOCATION:CE 1 2
STATUS:CONFIRMED
END:VEVENT
END:VCALENDAR