Foundations of Deep Learning

This unit introduces core concepts, tools, and math needed to understand and begin building deep learning models. Students will learn key terminology, use Python and common frameworks to implement simple networks, and survey model families and real-world considerations that inform downstream units.

Learning Objectives

• Analyze the distinctions between deep learning, machine learning, and traditional programming by comparing use cases, strengths, and limitations of each approach
• Apply Python and at least one deep learning framework to implement, train, and evaluate a simple perceptron and a small feedforward neural network on a basic classification dataset
• Demonstrate foundational mathematical concepts (linear combinations, gradients, and basic probability) by deriving and computing the operations that underlie training and inference in neural networks
• Evaluate the applicability of CNNs, RNNs, generative and adversarial methods to different problem types and identify ethical, human-centered, and deployment considerations when selecting an approach