Machine Learning and Imaging Lecture 6: Ingredients for Machine Learning

In this lecture, Dr. Horstmeyer presents the general goals and outline of a machine learning algorithm. Building upon Lecture 5's description of computational optimization, he first describes the setup of an optimization problem in terms of a dataset, cost function, and optimization variable of interest. He then transitions this framework to the case of machine learning, where the goal is to optimize a mapping between a dataset and known labels for that dataset during algorithm training. He then presents a first example machine learning problem that considers how to classify images of handwritten numerical digits (the MNIST dataset classification problem). Additional course material is available at deepimaging.github.io
#machinelearning #cameras #medicalimaging #ai #tensorflow