[Colloquium] Reminder - Jacob Williams MS Presentation/Feb 3, 2023

[Megan Woodward]

This is an announcement of Jacob Williams's MS Presentation
===============================================
Candidate: Jacob Williams

Date: Friday, February 03, 2023

Time:  3 pm CST

Remote Location: https://uchicago.zoom.us/j/98622767991?pwd=eDBmRTdQd0N3ZWVsbjJ5T1hWdmk2UT09<https://urldefense.com/v3/__https://uchicago.zoom.us/j/98622767991?pwd=eDBmRTdQd0N3ZWVsbjJ5T1hWdmk2UT09__;!!BpyFHLRN4TMTrA!_km8JSORgHOxp1N3qL3S8OfE8B7HpndyiShEfComfW9vAA33TYCHNzr7nvSWp-N2ZAaawTOOXGdkZHqO7P9bfEqSyCsqN6d7zo97$>  Meeting ID: 986 2276 7991 Passcode: 287940

Location: JCL 298

M.S. Paper Title: Rapid Prediction of Full Spin Systems using Uncertainty-Aware Machine Learning

Abstract: Accurate simulation of solution NMR spectra requires knowledge of all chemical shift and scalar coupling parameters, traditionally accomplished by heuristic-based techniques or ab initio computational chemistry methods. Here we present a novel machine learning technique which combines uncertainty-aware deep learning with rapid estimates of conformational geometries to generate Full Spin System Predictions with UnCertainty (FullSSPrUCe). We improve on previous state of the art in accuracy on chemical shift values and are able to predict all scalar coupling values, unlike previous GNN models. Our uncertainty quantification shows a strong, useful correlation with accuracy, with the most confident predictions having significantly reduced error. We can also correctly handle stereoisomerism and intelligently augment experimental data with ab initio data through disagreement regularization to account for deficiencies in training data.

Advisors: Eric Jonas

Committee Members: Ian Foster, Eric Jonas, and Rebecca Willett

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.cs.uchicago.edu/pipermail/colloquium/attachments/20230203/108843a0/attachment-0001.html>