Class Central is learner-supported. When you buy through links on our site, we may earn an affiliate commission.

Machine Learning for Core Engineering Disciplines

NPTEL via Swayam

Go to class Write review

ABOUT THE COURSE:This course will introduce students and practitioners of core engineering disciplines to the fundamentals and applications of machine learning. The course will cover the following material: introduction to data science and machine learning, examples of supervised/unsupervised/reinforcement learning, motivation for machine learning with examples derived from various core engineering disciplines; introduction to Python, scientific computing packages (NumPy, SciPy, Matplotlib), and simple ML packages (Scikit-learn, TensorFlow); linear and nonlinear regression, confidence intervals and goodness of fit, loss functions, gradient descent algorithm, overfitting/underfitting, regression/classification learning; clustering, singular-value decomposition, and principal component analysis; decision trees and ensemble methods, boosting and bagging techniques, random forests, gradient-boosted machine learning, support vector machines, Gaussian process regression; hyperparameter tuning and cross validation; introduction to neural networks and deep learning; feed-forward, convolutional, and recurrent neural networksINTENDED AUDIENCE: Core Engineering DisciplinesINDUSTRY SUPPORT: Several companies working in core engineering disciplines

Syllabus

Week 1: Introduction to data science and machine learning; differences between supervised, unsupervised, and reinforcement learning; introduction to probability and statistics; sample and population properties; covariance and correlation matrix
Week 2:Linear regression; model parametrization and fitting; coefficient of determination
Week 3:Logistic regression; implementation of models in Python
Week 4:Overfitting and underfitting; bias-variance tradeoff dealing with overfitting via regularization (ridge regression and LASSO)
Week 5:Confidence intervals and hypothesis testing
Week 6:Nonlinear regression; loss functions; gradient descent algorithm and its variations; Cross validation and hyperparameter tuning
Week 7:Unsupervised learning; singular value decomposition; principal component analysis; clustering algorithms
Week 8:Decision tress; ensembling; random forests
Week 9:Bagging and boosting; gradient-boosted decision trees
Week 10:Introduction to neural networks
Week 11:Feed-forward and convolutional neural networks
Week 12:Recurrent neural networks