This talk covers the fundamentals of machine learning and data science, focusing on material science applications. The talk is for a general audience, attempting to introduce basic concepts such as linear regression, supervised learning with neural networks including forward and back propagation, and concepts of underfitting and overfitting. The talk lists a set of handy resources towards the end and builds on to a hands-on demonstration using the nanoHUB tool Machine Learning for Materials Science: Part 1.

Saaketh Desai is a PhD student in Prof. Alejandro Strachan's research group at the School of Materials Engineering, Purdue University. Saaketh's research interests lie in use of novel computational techniques to improve the predictive power of molecular dynamics simulations, extending the accuracy and timescales achievable via these simulations. Saaketh is also interested in the use of machine learning approaches to enhance physics-based atomistic modeling approaches. Saaketh received his undergraduate degree in Metallurgical Engineering from the Indian Institute of Technology, Roorkee.

Nanotechnology Student Advisory Council

Researchers should cite this work as follows:

• Saaketh Desai (2020), "Data Science and Machine Learning for Materials Science," https://nanohub.org/resources/32274.

  BibTex | EndNote

1. data science
2. joes picks
3. linear regression
4. machine learning
5. material properties
6. materials science
7. NCN Group - Machine Learning
8. NCN Group - Materials Science
9. neural networks
10. supervised learning