Applied Deep Learning

Catalog description

This course provides a broad, hands-on exploration of applied deep learning across modern problem domains including computer vision, sequence modeling, audio and speech processing, and multimodal learning. Students learn to identify different machine learning problem types and select the most appropriate deep learning architectures for each. A major emphasis is placed on developing proficiency with PyTorch-building custom neural networks, writing training loops, managing data pipelines, leveraging GPUs, and debugging model behavior. Students apply rigorous experimental design and evaluation methods while working with real datasets and state-of-the-art models. The course culminates in an applied final project where students apply and evaluate techniques learned throughout the semester