BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Memento EPFL//
BEGIN:VEVENT
SUMMARY:Progress towards leveraging Machine Learning for Organic Synthesis
DTSTART:20231010T160000
DTEND:20231010T170000
DTSTAMP:20260506T180601Z
UID:bdb25307c51ed56f689d2fdc08b342799fc259c39a8fec654666c70b
CATEGORIES:Conferences - Seminars
DESCRIPTION:Jules Schleinitz. Jules is currently a postdoctoral scholar at
  CalTech in the group of Sarah E. Reisman and a current member of the NSF 
 Center for Computer Assisted Synthesis. His research focuses on the develo
 pment of computational and machine learning tools for organic synthesis pl
 anning through mechanistic understanding. Jules graduated in 2022 from the
  Ecole Normale Supérieure in Paris. His PhD intitled “Machine learning 
 and Mechanistic Analysis” was supervised by Laurence Grimaud. Alongside 
 with his research activities\, Jules spent half of his PhD teaching chemis
 try at Ecole Normale Supérieure (Organic chemistry: lessons\, electrochem
 istry: tutorials and experimental sessions\, experimental projects for bac
 helor and master students.).\nPredicting experimental organic synthesis ou
 tcomes is highly desirable for the development and testing of new drug can
 didates and the optimization of process chemistry. Machine learning offers
  great opportunities to tackle these challenging problems. However\, due t
 o the costs of experimentation and reaction product characterization\, mod
 els trained on experimental results must perform in a low data regime. The
  discussion of reaction yield prediction on literature-extracted data foc
 used on Nickel-Catalyzed C–O Couplings: NiCOlit\, will show how the str
 ucture of published data impacts model performances. We will draw conclusi
 ons about how to design experimental datasets that are suited for modeling
 . Finally\, we will discuss efforts to build surrogate models for the pred
 iction of nitrogen-based ligands for Nickel catalysis and regioselectivity
  predictions based on the literature.
LOCATION:https://epfl.zoom.us/j/68447908297?pwd=OU5JUGJUSUhZc0ZNYjQ2WENvYl
 NRdz09
STATUS:CONFIRMED
END:VEVENT
END:VCALENDAR