MATH 474 Machine Learning II
A continuation of MATH 473. This course develops the theory and implementation of artificial neural networks. Topics include multi-layer perceptrons, back propagation, stochastic gradient descent, dropout, learning rate schedulers, convolutional neural networks, long short-term memory networks, transfer learning, and deep neural networks. Emphasis is placed on practical implementation of neural networks using software. Students spend a significant amount of time developing projects from start to finish: data acquisition and preparation, network architecture design, model training with validation and hyperparameter tuning, and testing to prevent overfitting.
Prerequisite
Grade of C or better in
MATH 473.