• View
• Edit
• History
• Print

Next

 

More advanced ML courses (only offered some years):

• CIS625: Computational Learning Theory
  • theory side of ML (CS version)
• STAT928: Statistical Learning Theory
  • theory side of ML (Stat version)
• CIS 700 Machine Learning and Economics
  • ranking and choice models
• Lots of opportunities for independent studies in different research groups
  • NLP, robotics, ML theory, medicine, ….

Statistics courses

• STAT542 - Bayesian Methods and Computation
  • Shane Jenson on Bayesian methods (EM, Gibbs sampling)
• Multivariate methods
  • Andreas Buja on multivariate methods (research oriented)
• STAT500 - Applied Regression and Analysis of Variance
  • never underestimate the value of really understanding regression, but STAT 500 is too basic for people who have CIS520
  • If you want the math side of regression, see STAT550 - Mathematical Statistics
• STAT553 - Machine Learning — probably too basic after 520
• STAT701 - Modern Data Mining — probably too basic after 520

Other related courses:

• CIS545 Big Data Analytics
  • “data-parallel” approach to scaling computation
• ESE605: Modern Convex Optimization
  • useful

Many courses go into detail on applications of machine learning to e.g. vision or NLP

• ESE 650 Learning in Robotics
  • Dan Lee — serious ML for Robotics
• CIS521: Intro to AI
  • much easier than 520, with a bit of overlap (and not much ML)
• CIS580: Machine Perception

Again: there are also many options to do research in various labs around Penn.

Back to Lectures