Additional Content
Unless otherwise indicated, a grade of C or higher is required for all prerequisite courses.
Introduces the theory and practice of deep learning. Topics include neural network architectures; activation functions; backpropagation; optimization (SGD, momentum, Adam); convolutional, recurrent, and transformer models; regularization and generalization; supervised and unsupervised learning; and model evaluation. Students implement and train models with an industry-standard framework to solve problems in computer vision, natural language processing, and sequence modeling. The course emphasizes reproducible experimentation, responsible/ethical use, and efficiency considerations (batching, initialization, mixed precision).