Overview
Explore advanced concepts in machine learning classification through this 35-minute lecture from the University of Central Florida's CAP5415 course. Delve into Support Vector Machines (SVM), beginning with an introduction and background on maximal-margin classifiers. Examine the robustness of max-margin approaches and address non-separable cases. Learn about support vector classifiers for both separable and overlapping scenarios. Analyze the disadvantages of linear decision surfaces and discover the advantages of non-linear alternatives. Conclude by investigating nonlinear SVMs and their application in classification tasks. This comprehensive lecture provides a thorough understanding of SVM principles and their practical implementation in machine learning.
Syllabus
Intro
SVM - background
Maximal-Margin Classifier
Is Max-Margin Robust?
Non-Separable Case
Support Vector Classifier-Separable
Support Vector Classifier-Overlap
SV Classifier
Disadvantages of Linear Decision Surfaces
Advantages of non-linear Decision Surfaces
Nonlinear SVMS
SVM Classifier
Taught by
UCF CRCV