ME8813 - Machine Learning Fundamentals for Mechanical Engineering

Instructors: Prof. Yan Wang and Prof. Surya Kalidindi

Topics

Introduction to AI and ML in Engineering

Searching Algorithms(BFS, DFS, greedy, global optimization)

Constraint Satisfaction Problems & Applications in Design

Uncertainty and Probabilistic Reasoning (probability theory, Dempster-Shafer, Bayesian belief network, MCMC)

Markov Models (Markov chain, hidden Markov model)

Training Markov Models with Applications in Manufacturing (E-M algorithm, Viterbi algorithm, Baum-Welch Algorithm)

Supervised Learning (Bayesian regression, LASSO, Gaussian process)

Artificial Neural Networks, Deep Neural Networks & Applications in Engineering

Unsupervised Learning (k-means, self-organizing map, linear discriminant analysis, linear and nonlinear dimensionality reductions)

Reinforcement Learning and Active Learning (Markov decision making, Bayesian optimization) & Applications in Control

Example Projects in Previous Years

Data-Driven Approach to Model Electric Vehicle Drivetrains	The Future of Airport Pushbacks Use of Reinforcement Learning in Ground Support Equipment
The Effectiveness of Various Sensors for Tool Condition Monitoring of Flank Wear for Predictive Maintenance on Milling Machines	Methods in Image Classification of Manufactured Parts Through Training on 3D Models
Identifying Notable COVID-19-associated Metabolites Through Clustering Analysis of Healthy and COVID-positive Blood Samples	Exploring Convolutional Neural Networks as a Supplement to Manufacturing Visual Inspection
Evaluating Deep Convolutional Neural Networks for Remaining Useful Life Estimation of Turbofan Engines	Operating Parameter Prediction of a Hydrogen Electrolyzer using EIS: A machine learning approach
Machine Learning Models for Human Activity Recognition	Detecting Arrhythmias in Fetuses using Long Short-Term Memory Architecture on Non-Invasive Fetal Electrocardiograms
Machine Learning Informed Biological Torque Prediction in Industrial Lifting Tasks	Toward a Cyber-secure Autonomous Control System for Nuclear Power Plants
Machine Learning Interatomic Potentials for Molecular Dynamics Simulations	A CNN model for effective property prediction of 2-phase microstructures with varying contrast ratios of elastic properties
Predicting Crystal Symmetry: Missing values and Imbalanced Datasets	Bandgap Enhancement via Gaussian process-based Optimization of Piezoelectric Metamaterials
Interference Detection in Laser Powder Bed Fusion Recoater Blade Using Machine Learning	Transferability of Geometry Predictions in WAAM
Predicting the State of a Manufacturing Process with Hidden Markov Models	Changepoint Detection for Porosity of Additively Manufactured Metal Parts
Machine Learning Approaches for Considering End-of-Life Electric Vehicle Battery Pre-Processing Facility Locations	Predicting the Water Level of the Lake Mead Reservoir using Climate and Freshwater Demand Data