webinar register page

Advanced topics in using Exabyte.io platform
Host: Timur Bazhirov, Ph.D.

Presenter(s): James Dean, Ph.D.

Title: Interpretable Machine Learning for Materials R&D

Join our team expert, Dr. James Dean, as he introduces our recent work on Interpretable Machine Learning: a recent collaborative approach to building data-driven predictive models for materials.


# Abstract

Fueled by the widespread adoption of Machine Learning and the high-throughput screening of materials, the data-driven approach to materials design has asserted itself as a robust and powerful tool for the in-silico prediction of materials properties. When training models to predict material properties, researchers often face a difficult choice between a model's interpretability or its performance. We study this trade-off by leveraging four different state-of-the-art ML techniques: XGBoost, SISSO, Roost, and TPOT for the prediction of structural and electronic properties of perovskites and 2D materials. We then assess the future outlook of the continued integration of ML into materials discovery and identify key problems that will continue to challenge researchers as the size of the literature's datasets and complexity of models increases. Finally, we offer several possible solutions to these challenges with a focus on retaining interpretability and share our thoughts on magnifying the impact of ML on materials design.

# What will be covered?

Background

The need for FAIR data
Overview of several recent ML techniques
Motivation and Aims
The Approach

## Datasets:
NOMAD
2DMatPedia
Others

## Feature Engineering
XenonPy
Matminer
Other


## The Applications

Predicting Volume of 3D Perovskites
Predicting Electronic Bandgaps of 2D Materials
Predicting Exfoliation Energies of 2D Materials

## Summary and Q&A
Please choose only one webinar to attend.
Time shows in
* Required information
Loading