30

per person /
$

Free

Home Courses

Artificial Intelligence in Materials Science

Artificial Intelligence in Materials Science

Teacher

Dr. Satyanarayan Dhal

Category

Skill Courses

Course Attendees

Still no participant

Course Reviews

Still no reviews

Code (Credit): CUTM 3164 (0-2-2)

Course Objectives:

  • To understand the fundamentals of machine learning and its applications in materials science.
  • To apply machine learning algorithms to solve materials science problems, evaluate their performance, and interpret their results.
  • To gain hands-on experience using Python libraries for machine learning, as well as an understanding of the limitations and challenges of machine learning in materials science.
  • This course exposes students to a complete example of a machine learning (ML) approach applied to a materials science dataset to solve a research topic.
  • Learning Outcomes:
  • Understand the basic concepts of machine learning and its applications in materials science.
  • Develop skills to use Python libraries for machine learning, such as scikit-learn, Pymatgen, and graphviz.
  • Gain hands-on experience in applying machine learning techniques to solve materials science problems.
  • Learn how to evaluate the performance of machine learning models and interpret their results.
  • Develop a critical understanding of the limitations and challenges of machine learning in materials science.

Prerequisites:

  • Basic knowledge of programming in Python is recommended.
  • Understanding of calculus
  • Basic knowledge of materials science is recommended but not required.

Course Syllabus

Module I: Setup (In Workshop Mode) (2 Hours)

Module II: Data Inspection: (4 Hours)

Practice 01: Importing the Dataset, [1 Hour]

Practice 02: Data Cleaning, [1 Hour]

Practice 03: Evaluation of Data availability. [1 Hour]

Module III: Feature Generation (4 Hours)

Practice 04: Materials Simulation Toolkit for Machine Learning configuration [2 Hrs]

Practice 05: Validating Compositional average features [2 Hrs]

Module IV: Feature Engineering (4 Hours)

Practice 06: Removal of Constant Columns and highly correlated features; [1 Hrs]

Practice 07: Feature Normalization [2Hrs]

Practice 08: Establishment and Evaluation of Train/Test Split [1 Hr]

Module V: Setup for Model Evaluation (4 Hours)

Practice 09: Fitting the decision tree model, [1 Hr]

Practice 10: evaluating model performance on training and test data;

Practice 11: Visualizing and interpreting the decision tree model. [2 Hrs]

Module VI: Hyperparameter Optimization (4 Hours):

Practice 12: Setting up a cross-validation scheme; [1 Hr]

Practice 13: Setting up a grid search; [1 Hr]

Practice 14: Visualizing the learning curve; [1 Hr]

Practice 15: cross-validation performance [1 Hr]

Module VII: Make Predictions (2 Hours)

Practice 16: Make Predictions [2 Hrs]

 

Our Main Teachers