Applied Machine Learning 2-day course
From RAD Lab
Contents |
Date, Time and Place
August 23-24, 9:00-5:00 both days, 306 Soda Hall
Registration is required as space is limited. Email Cecilia Pracher to register and/or if you need a parking pass; try to carpool as these are limited. Or, from the Downtown Berkeley BART station, walk to Soda Hall in 15 minutes, or take the UC Berkeley Perimeter bus, which runs every 12 minutes starting at :00 from in front of the Bank of America outside the BART station.
Video
Schedule
Thursday, Aug 23
- 9:00-10:00 Classification (Simon Lacoste-Julien)
- 10:00-11:00 Regression (Kurt Miller)
- 11:00-11:20 Break
- 11:20-12:30 Feature Selection (Alex Shyr)
- 12:30-1:30 Lunch
- 1:30-2:30 Diagnostics (Gad Kimmel)
- 2:30-3:30 Clustering (Junming Yin)
- 3:30-3:45 Break
- 3:45-5:00 Graphical Models (Mike Jordan)
Friday, Aug 24
- 9:00-10:00 Linear Dimensionality Reduction (Percy Liang)
- 10:00-10:30 Break
- 10:30-11:30 Manifolds and Visualization (Fei Sha)
- 11:30-12:30 Lunch
- 12:30-1:30 Structured Classification (Simon Lacoste-Julien)
- 1:30-2:30 Reinforcement Learning (Peter Bodik)
- 2:30-3:00 Break
- 3:00-5:00 Nonparametric Bayesian Models (Mike Jordan)
Notes
We will not be making any strong assumptions regarding your background in statistics, probability or machine learning---the course will be introductory and self-contained. That said, we want to move fairly rapidly, and it will be helpful if we can assume all attendees have at least some minimal exposure to basic ideas of probability (expectation; independence; conditional independence), statistics (distributions such as the binomial, Gaussian and Dirichlet; the exponential family; maximum likelihood) and linear algebra (eigenvectors; SVD; vector derivatives). These slides overview some of these basic ideas.
