Deep Neural Networks

Introduction to biological and artificial neurons, learning from data, artificial neural networks, and non-linear activation functions, error backpropagation and restricted Boltzman machine algorithms, deep vs. shallow learning, data augmentation, the theory of generalization, convolutional neural networks, recurrent neural networks, deep unsupervised and reinforcement learning, parallel computing for AI (GPU computing, CuDNN, etc.), and application areas of deep learning – speech recognition, images, vision, etc.

Hours: XYZ where X = Lecture, Y = Lab, Z = Credit
All hours are per week.
3 Lab hours constitute 1 credit hour
1 credit hour implies 1 lecture of 50mins per academic week. 16 weeks in total.
Pre-Requisite courses are courses required to be completed before this course may be taken
Co-Requisite courses are courses required to be taken along with this course