How do we use mathematical thinking to design better machine learning methods? This course focuses on developing mathematical tools for answering these questions. This course will cover fundamental concepts and principled algorithms in machine learning. We have a special focus on modern large-scale non-linear models such as matrix factorization models and deep neural networks. In particular, we will cover concepts and phenomenon such as uniform convergence, double descent phenomenon, implicit regularization, and problems such as matrix completion, bandits, and online learning (and generally sequential decision making under uncertainty).

Topics Include

• Concentration inequalities
• Generalization bounds via uniform convergence
• Non-convex optimization
• Implicit regularization effect in deep learning
• Unsupervised learning
• Domain adaptations

Note: This course is cross listed with CS229M.

What You Need to Succeed

• A conferred Bachelor’s degree with an undergraduate GPA of 3.3 or better
• Linear algebra (MATH51 or CS205L), probability theory (STATS116, MATH151, or CS109), and machine learning (CS229 or STATS315A) or equivalents

What You Need To Get Started

Before enrolling in your first graduate course, you must complete an online application.

Don’t wait! While you can only enroll in courses during open enrollment periods, you can complete your online application at any time.

Once you have enrolled in a course, your application will be sent to the department for approval. You will receive an email notifying you of the department's decision after the enrollment period closes. You can also check your application status in your mystanfordconnection account at any time.

Learn more about the graduate application process.

How Much It Will Cost

Learn more about tuition and fees.