Algorithmic Machine Learning and Data Science

Course Description

This course gives a behind-the-scenes look into the algorithms and computational methods that make machine learning and data science work at large scale. How does a service like Shazam match a sound clip to a library of 10 million songs in under a second? How do scientists find patterns in terabytes of genetic data? How can we efficiently train neural networks with millions of parameters on millions of labeled images? We will address these questions and others by studying advanced algorithmic techniques like randomization, approximation, sketching, continuous optimization, spectral methods, and Fourier methods. Students will learn how to theoretically analyze and apply these techniques to problems in machine learning and data science. They will also have the opportunity to explore recent research in algorithms for data through a final project and optional reading group. This course is mathematically rigorous and is intended for graduate students or strong, advanced undergraduates. | Knowledge of machine learning (equivalent to CS-UY 4563, CS-GY 6923, or ECE-GY 6143), algorithms (equivalent to CS-UY 2413, CS-GY 6033, or CS-GY 6043), and linear algebra (equivalent to MA-UY 2034, 3044, or 3054). Prerequisite: Graduate Standing. Prerequisites: Graduate Standing.

Code

CS-GY 6763

Credits

3

Subject

Computer Science

Programs That Include This Course

• Doctoral Urban Systems (PhD)
• Master's Computer Science Tandon (MS)
• Master's Emerging Technologies (MS)