Course Syllabus: AI Methods in Advanced Materials Research

• When: Starting January 19th, 2024, Every Friday (during seminar days), 9:30 am, lasting 45 minutes per session (with additional time afterwards for questions), spanning around 10 weeks.

• Where: JSNN Auditorium.

• Course Website: CLICK HERE

• Description: The course integrates AI methods into graduate-level research, focusing on material research. Topics covered include neural networks, large language models, machine learning, featurization, hyperparameter tuning, and introductory-level advanced analysis.

• Instructor: Jared Keith Averitt

  • Prefered Email: [email protected]
  • UNCG Email: [email protected]
  • Developed with guidance from Dr. Starobin and Dr. Ignatova.

• Requirements: Bring your laptop. You will not need to install anything to start. We will be using Jupyter notebooks through Google Colab. You will need a google account (your UNCG gmail will not work). The course begins with Python basics, progressing to large language models and convolutional neural networks. Moderate coding understanding (e.g., Matlab) is beneficial; extra Python learning resources will be provided.

• Python Crash Course (Additional Resources to get you up to date with Python)

• 1.5 Hr -- Very Short Crash Course

• 9h -- Basically a class on Python – Starts from 0 Python Knowledge all the way to Pandas Data Frames

  • Both Recomended by Anna Sheets

• Expected Learning Outcomes: Implementation of AI in research, handling experimental results in data frames using Pandas and Python, utilizing PyTorch, scikit-learn, and featurization tools.

• Assignments: Projects follow each lecture to test competency. Solutions are released before the subsequent lecture. Collaborative work is encouraged. Self-graded assignments with opportunities for guidance and feedback.

• Covered Topics: (Subject to change)

  Week	Date	Topic	Assignment	Solutions
  1	01/19/2024	Intro to AI in Python	NumPy_Assignment_1	Assignment_1_Solutions
  2	01/26/2024	Machine Learning Using Sklearn	ML_DF_Assignment_2	Assignment_2_Solutions
  3	02/09/2024	Multi Layer Perceptrons
  4	02/16/2024	PyTorch	Non-linear Classifier	Assignment 4 Solutions
  5	02/23/2024	Convolutional Neural Networks
  6	03/01/2024	Material Databases
  7	03/15/2024	Atomic Simulation Environment
  8	04/05/2024	TorchANI
  9	04/12/2024	LLM
  10	04/19/2024	LLM

• Survey Link: Complete initial survey

• Lecture Notes Books on Computational Material Science : [https://www.csfreelist.com/materialScBooksLinks.php]

• Libraries and Tools for Computational Material Science : [https://matgl.ai] , [https://matbench-discovery.materialsproject.org], [www.TorchANI.]