ME644
|
MACHINE LEARNING FOR ENGINEERS
|
|
|
Course contents:
Mathematical preliminaries, python programming, simple/multiple linear regression, nonlinear regression, logistic regression, k-nearest neighbours, perceptrons, random forest, naïve Bayes, support vector machines, artificial neural network, clustering, dimensionality reduction
Lecturewise Breakup (Based on 50 min per lecture)
Sl. No. | Topic | Contents | Lecture |
1. |
Introduction |
Various learning paradigms, definitions, examples |
1 |
2. |
Programming |
Programming in python, libraries: scientific computing, machine learning, plotting |
2 |
3. |
Mathematics for machine learning |
Linear algebra and vector calculus: Vector space, vector-matrix operations, norm, eigenvalue and eigenvectors, matrix decompositions, differential calculus of vectors |
3 |
Optimization: gradient-based techniques, metaheuristic techniques, numerical implementation |
3 |
Statistics and probability: Probability distributions, hypotheses testing, Bayes’ theorem |
3 |
4. |
Supervised learning |
Linear/nonlinear regression, overfitting, regularization, logistic regression, naive Bayes, k-NN, decision tree, random forest, maximum likelihood, support vector machine, applications in mechanical engineering |
15 |
5. |
Unsupervised learning |
Singular value decomposition, principal component analysis, clustering, applications in mechanical engineering |
8 |
6. |
Artificial neural network |
Single- and multi-layer networks, activation, backpropagation, stochastic gradient descent, physics-informed neural network, applications in mechanical engineering |
5 |
Total |
40 |
References:
-
Machine Learning for Engineers, R. G. McClarren, Springer
-
A First Course in Machine Learning, S. Rogera, M. Girolami, CRC Press
-
Machine Learning, Z-H. Zhou, Springer
-
An Introduction to Statistical Learning, G. James et al., Springer
-
Data-Driven Science and Engineering, S. L. Brunton, J. L. Kutz, Cambridge Uni. press
-
Probabilistic Machine Learning for Civil Engineers, J-A. Goulet, MIT Press
-
Machine Learning Refined, 2nd ed., J. Watt et al., Cambridge University press
-
Machine Learning, A. Lindholm et al., Cambridge University press
|