Introduction to Machine Learning for Mechanical Engineers
This course introduces Machine Learning (ML) concepts and methods to Mechanical Engineering (ME) students. The course starts with basic supervised and unsupervised learning methods, ensemble methods, the bias-variance tradeoff, and model selection. It then covers two topics in detail: Neural Networks as a key method in the modeling of mechanical systems, and Reinforcement Learning as a key method in robotics and cyber-physical systems. The course is structured around hands-on assignments that illustrate the use of ML in addressing ME problems. It is tailored to ME students as it builds on an engineering mathematics and probability foundation along with basic programming skills but does not assume a background in algorithms.
Topics
- Distinguish between supervised and unsupervised learning methods, evaluate their suitability for solving mechanical engineering problems, and apply the appropriate methods to make inferences.| Apply model selection strategies and analyze the bias-variance tradeoff to select appropriate ML models for engineering datasets.| Implement neural network methods to model mechanical systems and interpret their output.| Implement reinforcement learning methods for decision-making systems.| Identify and critically evaluate the implications of machine learning and AI technologies in mechanical engineering practice, including their ethical, societal, and professional impacts.
Official roster learning outcomes
Prerequisites & course information
MATH 2940; CS 1110 or CS 1112; ENGRD 2700; or equivalents.
Sources
- Course catalogue ↗2026-2027
- Catalogue entry ↗
Catalogue variants
- Introduction to Machine Learning for Mechanical Engineers · 3.0 · Undergraduate