Mini Course: Practical Machine Learning for Polymer Chemistry

Machine learning (ML) is becoming an increasingly useful tool for accelerating chemical and materials discovery. This mini-course will provide a practical introduction to applying machine learning to polymer chemistry, with an emphasis on small chemical datasets and experimentally relevant chemistry problems.
The course will introduce molecular descriptors, predictive modeling, model validation, Bayesian optimization, and interpretable machine learning. Examples from polymer property prediction and polymerization catalyst discovery will be used to illustrate how machine learning can help guide experiments and identify new chemical design principles.
The course is intended for students and researchers with backgrounds in chemistry, materials science, chemical engineering, or related fields. No prior machine-learning experience is required.

This is lecture 1 of a 5-part series.

(1)  Machine Learning for Chemists: Concepts and Opportunities
Introduction to machine learning, chemical datasets, regression and classification, and applications in polymer science.
(2-3) Representing Molecules and Polymers for Machine Learning
Molecular descriptors, fingerprints, steric/electronic descriptors, DFT-derived descriptors, and some basic DFT computation using Gaussian / ORCA.
(4) Building and Evaluating Predictive Bayesian Optimization Models 
Regression models, Gaussian processes, cross-validation, model evaluation, uncertainty, common pitfalls, experiment selection, exploration vs. exploitation, acquisition functions, and iterative ML–experiment workflows.
(5) Interpretable ML and Building an Autonomous Discovery Workflow
SHAP analysis, extracting chemical insights from ML models, integrating ML with computation and experiments, and closed-loop materials discovery.

Please REGISTER here if you plan on coming...