ACS 3520 Deep Learning (DS 2.2)
This course explores the cutting edge of data science: deep learning. This course compares and contrasts several neural network architectures for solving problems across different domains including image classification, and face recognition. Students build and train a perceptron which is the basic computational unit of artificial neural networks. The course will cover the limitations of perceptrons and how non-linear activation functions enhance their power, how to combine many perceptrons to construct feed-forward neural networks and how to program a training algorithm using error backpropagation and gradient descent. Students will use cutting-edge software libraries and tools including Keras and TensorFlow to construct large-scale networks with relatively little code. Prerequisites:
ACS 3510 (DS 2.1)