Convex Optimization: Spring 2015

Machine Learning 10-725
(cross-listed as Statistics 36-725)

Instructor: Ryan Tibshirani (ryantibs at cmu dot edu)

Mattia Ciollaro (ciollaro at cmu dot edu)
Junier Oliva (joliva at cs dot cmu dot edu)
Nicole Rafidi (nrafidi at cs dot cmu dot edu)
Veeranjaneyulu Sadhanala (vsadhana at cs dot cmu dot edu)
Yu-Xiang Wang (yuxiangw at cs dot cmu dot edu)

Course assistant: Mallory Deptola (mdeptola at cs dot cmu dot edu)

Lecture times: Mondays and Wednesdays 10:30-11:50am, Hamburg Hall 1000

Office hours:
RT: Mondays 12-1pm, Baker 229B
MC: Mondays 1-2pm, Wean 8110
JO: Fridays 1-2pm, GHC 8229
NR: Tuesdays 1-2pm, common area on 8th floor of GHC
YW: Wednesdays 2-3pm, common area on 8th floor of GHC

Syllabus: here

Scribing: Sign up for scribing here, and find the Latex scribing template here

Discussions: Class discussions will be run through Piazza, linked here

Go to:   Schedule | Assignments | Miscellaneous


Here is the estimated class schedule. It is subject to change, depending on time and class interests.

Theory I: Fundamentals
Mon Jan 12 Introduction Slides, notes, video Quiz Scribe
Wed Jan 14 Convexity I: Sets and functions Slides, notes, video Quiz
Mon Jan 19 (MLK day, no class)
Wed Jan 21 Convexity II: Optimization basics Slides, notes, video Quiz
Mon Jan 26 Canonical problem forms Slides, notes, video Quiz
Algorithms I: First-order methods
Wed Jan 28 Gradient descent Slides, notes, video Quiz Hw 1 due (Thu)
Mon Feb 2 Subgradients Slides
Wed Feb 4 Subgradient method Slides
Mon Feb 9 Proximal gradient descent, acceleration Slides
Wed Feb 11 Numerical linear algebra review Project milestone 1 due
Theory II: Optimality and duality
Mon Feb 16 The dual problem Hw 2 due (Tue)
Wed Feb 18 KKT conditions
Mon Feb 23 Duality correspondences
Wed Feb 25 Practice with the KKT conditions
Algorithms II: Second-order methods
Mon Mar 2 Newton's method Hw 3 due (Tue)
Wed Mar 4 Log-barrier method
Mon Mar 9 (Spring break, no class)
Wed Mar 11 (Spring break, no class)
Mon Mar 16 Primal-dual interior point method
Wed Mar 18 Back to LPs, QPs, SDPs Hw 4 due (Thu)
Mon Mar 23 Recap: putting it all together
Wed Mar 25 Midterm Midterm
Mon Mar 30 Case study: lasso-type problem Project milestone 2 due
Wed Apr 1 Case study: ???
Special topics
Mon Apr 6 Dual methods and ADMM Hw 5 due (Tue)
Wed Apr 8 Coordinate descent
Mon Apr 13 Conditional gradient method
Wed Apr 15 Proximal and projected Newton methods
Mon Apr 20 Exact path-following methods or screeing rules?
Wed Apr 22 Large-scale optimization Hw 6 due (Thu)
Mon Apr 27 Little test Little test
Wed Apr 29 (Work on projects, no class)



All assignments are to be turned into Mallory Deptola, in 8001 GHC, unless specified otherwise. Top