Workshop Description

Through a series of rapid surveys, we will present an overview of recent topics in deep learning and machine learning with particular relevance for practitioners. Areas will include SVMs and kernels, variational autoencoders (VAEs) and dimensionality reduction, transfer learning, representation learning, and weakly supervised / semi-supervised / self-supervised learning. This workshop will assume a familiarity with basic concepts from both machine learning and deep learning as taught in the introductory workshops on those topics, but it will not assume a deep statistical background. Prior exposure to neural networks is highly recommended.

About the Instructors

Sherrie Wang graduated from Stanford in 2021 with a PhD in Computational and Mathematical Engineering and is now a Ciriacy-Wantrup Postdoctoral Fellow at UC Berkeley. She works on developing machine learning methods for remote sensing applications, especially in settings where ground truth labels are scarce. These methods are then applied to problems in sustainable agriculture and development, such as mapping where crops are grown in developing countries.

Alexander Ioannidis earned his Ph.D. in Computational and Mathematical Engineering and Masters in Management Science and Engineering both at Stanford University. He is a research fellow working on developing novel machine learning techniques for medical and genomic applications in the Department of Biomedical Data Science at Stanford. Prior to this he earned a bachelors in Chemistry and Physics from Harvard, an M.Phil from the University of Cambridge and conducted research for several years on novel superconducting and quantum computing architectures. In his free time, he enjoys sailing.

Pre-workshop Checklist

1. Sign up for Piazza (class code icme). We will be using Piazza to answer questions during the workshop.
2. You should have received a welcome email with the Zoom link and password. Please email us (tanishj [at] stanford [dot] edu) if you haven’t.
3. Familiarize yourself with the schedule and see you Monday August 16th at 8:00 am PT!

Schedule and Slides

Monday August 16th

• Session 1 (8:00 AM to 9:30 AM) [slides]
  • Introduction: Intermediate Topics in ML
  • Additional resources: Excellent free textbook with code examples (SVM example 9.6.2, ROC example 9.6.3); great free Convex Optimization textbook
• Session 2 (9:30 AM to 11:00 AM) [slides]
  • Transfer Learning
  • Generating Labels for Deep Learning

Tuesday August 17th

• Session 3 (8:00 AM to 9:30 AM) [slides]
  • Dimensionality Reduction
  • Variational Autoencoders (VAEs)
• Session 4 (9:30 AM to 11:00 AM) [slides]
  • Representation Learning and Self-Supervised Learning
  • Weakly Supervised Learning