Syllabus - Machine Learning for Automobile Applications (EV 504 (a))

Electric Vehicles

Machine Learning for Automobile Applications (EV 504 (a))

V Semester

UNIT I

Bayesian Decision Theory and Normal Distribution

Machine perception - feature extraction - classification, clustering, linear and logistic regression – Types of learning - Bayesian decision theory - classifiers, discriminant functions, and decision surfaces -univariate and multivariate normal densities - Bayesian belief networks.

UNIT II

Classification Algorithms

Perceptron and backpropagation neural network - k-nearest neighbor rule. Support vector machine: multicategory generalizations – Regression Decision trees: classification and regression tree – random forest.

UNIT III

Component Analysis and Clustering Algorithms

Principal component analysis - Linear discriminant analysis - Independent component analysis. K-means clustering - fuzzy k-means clustering – Expectation-maximization algorithm-Gaussian mixture models –auto associative neural network.

UNIT IV

Supervised and Unsupervised

Convolution neural network (CNN) - Layers in CNN - CNN architectures. Recurrent Neural Network - Applications: Speech-to-text conversion-image classification time series prediction.

UNIT V

Combining Multiple Learners

Generating diverse learners - model combination schemes - voting - error-correcting output codes -bagging - boosting - mixture of experts revisited - stacked generalization - fine-tuning an ensemble –cascading

Course Objective

This course aims to provide the required skill; 1. To introduce the fundamental concepts of machine learning and its applications 2.To learn the classification, clustering and regression based machine learning algorithms 3. To understand the deep learning architectures 4.To understand the methods of solving real life problems using the machine learning techniques 5. To understand the multiple learners, boosting and stacked generalization

Course Outcome

After completion of this course, students will be able to; 1. Understand the basic concepts of Bayesian theory and normal densities 2. Implement different classification algorithms used in machine learning 3. Implement clustering and component analysis techniques 4. Design and implement deep learning architectures for solving real life problems 5. Combine the evidence from two or more models/methods for designing a system

Practicals

Reference Books

  • R. O. Duda, E. Hart, and D.G. Stork, “Pattern Classification”, Second Edition, John Wiley & Sons, Singapore, 2012.

  • Francois Chollet, “ Deep Learning with Python”, Manning Publications, Shelter Island, New York, 2018.

  • Ethem Alpaydin, “Introduction to Machine Learning”, 3rd Edition, MIT Press, 2014.

  • C. M. Bishop, “Pattern Recognition and Machine Learning”, Springer, 2006.

  • Kevin P. Murphy, “Machine Learning: A Probabilistic Perspective”, MIT Press, 2012.

  • Navin Kumar Manaswi, “Deep Learning with Applications using Python”, A press, New York, 2018.