ECS 289G Scalable Machine Learning Fall 2015
Fall 2015


Tues/Thurs 10:30 am - 11:50 am




Instructor: Cho-Jui Hsieh
Office location: Mathematical Sciences Building (MSB) 4232
Email: chohsieh@ucdavis.edu
Office hours: Tues 1:30pm-2:30pm and by appointment
TA: Carlos Colman Meixner
Email: cecolmanmeixner@ucdavis.edu




Announcements


Overview

Course description

This is a course in machine learning for big data. The emphasis will be on developing scalable/parallel algorithms for various machine learning tasks. In addition to lectures on background material by the instructor, the course will also have paper presentations by students. Topics covered are expected to be: regression, classification, clustering, dimensionality reduction, matrix completion, parallel programming, optimization, etc. A substantial portion of the course will focus on research projects, where students will choose a well defined research problem.


Syllabus

    A high-level summary of the syllabus is as follows:
I. Supvervised learning: regression and classification
II. Optimization for Large-scale Machine Learning
III. Other Popular ML Problems

Grading Policy

Grades will be determined as follows:


Schedule

Date
Topic
Readings and links
Lectures
Assignments
Thurs 9/24
Course intro



lecture1_slides



Tues 9/29
Linear regression


lecture2_slides


Thurs 10/1
Intro to Optimization

Numerical Optimization (by Nocedal and Wright), section 1-3

lecture3_slides


Tues 10/6
Parallel Optimization



lecture4_slides


Thurs 10/8
Support Vector Machines



lecture5_slides


Tues 10/13
Optimization Solvers for Support Vector Machines
Pegasos: Primal Estimated sub-GrAdient SOlver for SVM
A dual coordinate descent method for large-scale linear SVM


lecture6_slides


Thurs 10/15
Matrix Completion
Maftrix factorization techniques for recommender systems


lecture7_slides


Tues 10/20
Matrix Completion


lecture8_slides
complexity


Tues 10/27
Multiclass and multilabel learning


lecture9_slides


Tues 11/3
Graph Algorithms


lecture10_slides


Thurs 11/5
Midterm Exam
lectures 2, 3, 5, 6



Thurs 11/12
Ranking


lecture11_slides