ECHE 589 - Machine Learning in Chemical Engineering
Instructor: Prof. Shengli Jiang
Term: Fall
Location: TBD
Time: Mondays, Wednesdays, and Fridays, 10:50 – 11:40 AM
Course Overview
This course provides a theoretical and practical introduction to machine learning (ML) methods and their applications in chemical engineering. After an overview of ML algorithms, we will focus on specific applications (e.g., QSPR modeling, molecular simulation, materials design, spectral and image analysis, and process control) to illustrate how ML methods widely adopted in Big Tech are applied to engineering problems. Through topical literature reviews, case studies, and programming-based assignments, students will engage with state-of-the-art methodologies and gain practical experience implementing ML algorithms in an engineering context.
Prerequisites
- ECHE 456: Computational Methods for Engineering Applications
- AND one of the following: CSCE 106 (Scientific Applications Programming) or CSCE 145 (Algorithmic Design I)
Schedule
| Week | Date | Topic | Materials |
|---|---|---|---|
| 1 | Week 1 | Course Introduction Overview of machine learning, course structure, and expectations. | |
| 2 | Week 2 | Overview of Machine Learning Fundamental concepts, branches of ML, and math review. | |
| 3 | Week 3 | Regression Linear and nonlinear regression, and model evaluation. | |
| 4 | Week 4 | Feature Engineering and Model Selection Data representation, feature engineering, and model selection techniques. | |
| 5 | Week 5 | Classification Logistic regression, support vector machines, random forests, and multi-class classification. | |
| 6 | Week 6 | Neural Networks Fundamentals Introduction to neural networks, TensorFlow/Keras/PyTorch, and gradient descent. | |
| 7 | Week 7 | Unsupervised Learning and Explainable AI Unsupervised methods, explainable AI, and best practices. | |
| 8-9 | Week 8–9 | Molecular-Level Machine Learning Molecular representations, featurization techniques, and molecular property prediction. | |
| 10-11 | Week 10–11 | Product-Level Machine Learning Bayesian optimization, active learning, and design of experiments. | |
| 12-13 | Week 12–13 | Process-Level Machine Learning Computer vision for process monitoring, fault detection, and diagnosis. |