Basic Introduction to ML
Jeffrey A. Fessler1
1University of Michigan, Ann Arbor, MI, United States
1University of Michigan, Ann Arbor, MI, United States
Synopsis
Basic introduction to machine learning.
Overview
This presentation will give a very basic introduction to machine learning, starting from basic definitions.Topics covered in the the differences between supervised and unsupervised learning, the importance of nonlinear operations, how one uses training data, validation data, and test data, and how one defines the learning process as an optimization problem. (Yes, there are a few equations.) The concepts are illustrated simple 1D and 2D examples.
The slides and Julia code for reproducing all the results in the presentation are available at this URL
https://tinyurl.com/ml2-18-jf
Acknowledgements
No acknowledgement found.References
Mehryar Mohri, Afshin Rostamizadeh, Ameet Talwalkar
Foundations of machine learning
MIT Press, 2nd Edition, 2018
https://mitpress.mit.edu/books/foundations-machine-learning
Figures
A simple neural network with two input features, one hidden layer, and a single output value (class label).
Neural-network based classifier for a two-dimensional input feature, with parameters learned from the training data shown.