Deep Multi-Task & Meta Learning | Stanford University CS 330

Explore state-of-the-art multi-task learning and meta-learning algorithms in this graduate-level Stanford course, with a focus on coding problems and a course project.

Introduction

This graduate-level course covers the setting where there are multiple tasks to be solved, and studies how the structure arising from multiple tasks can be leveraged to learn more efficiently or effectively. Topics include self-supervised pre-training, meta-learning methods, and curriculum and lifelong learning.

Highlights

• Covers state-of-the-art multi-task learning and meta-learning algorithms
• Focuses on coding problems that emphasize the fundamentals of these topics
• Includes in-person lectures, homework assignments, and a course project

Recommendation

This course is recommended for students interested in conducting research on multi-task learning and meta-learning. It assumes prior knowledge of machine learning and some familiarity with deep learning concepts.