ECS 171 Machine Learning (Winter 2018)
Winter 2018


Mon/Wed 6:10 pm - 7:30 pm




Instructor: Cho-Jui Hsieh
Office location: Mathematical Sciences Building (MSB) 4232
Email: chohsieh@ucdavis.edu
Office hours: Tuesday 1pm-2pm
TA: Pei-Hung (Patrick) Chen (phpchen@ucdavis.edu), Xuanqing Liu (xqliu@ucdavis.edu)
TA office hours: Thursday 10am-11am (Kemper 55)



Announcements
  1. Midterm exam: 2/21 (in class)

Overview

Course description

This course will roughly follow Learning from Data, which covers several important foundamental machine learning concepts and algorithms. Then we will introduce supervised learning algorithms (deep neural networks, boosting tress, SVM, nearest neighbors) and unsupservised learning algorithms (clustering, dimension reduction). Most pictures in lecture 1-12 are from the text book and slides by Prof. Abu-Mostafa and Prof. Lin.



Schedule

Date
Topic
Readings and links
Lectures
Assignments
Mon 1/8
Overview. The Learning Problem.

LFD 1.1, 1.2

lecture_1


Wed 1/10
Feasibility of Learning

LFD 1.3, 3.1

lecture_2


Wed 1/18
Linear Models I

LFD 3.2, 3.3

lecture_3


Mon 1/22
Gradient descent, SGD

LFD 3.3 and SGD

lecture_4


Wed 1/24
Error and Noise

LFD 1.4

lecture_5


Mon 1/29
Training versus testing

LFD 2.0, 2.1


lecture_6


Wed 1/31
Theory of generalization

LFD 2.1

lecture_7


Mon 2/5
VC Dimension

LFD 2.2

lecture_8


Wed 2/7
Nonlinear Transformation

LFD 3.4

lecture_9


Mon 2/12
Neural Network



lecture_10


Wed 2/14
Adversarial Examples in Deep Networks



lecture_11


Wed 2/21
Midterm Exam (cover material up to and including lecture 9)






Mon 2/26
Overfitting and Regularization

LFD 4.1, 4.2
lecture_12


Wed 2/28
Validation

LFD 4.3
lecture_13


Mon 3/5
SVM and Kernel methods


lecture_14


Wed 3/7
Decision tree, random forest, gradient boosting trees


lecture_15

Mon 3/11
Matrix factorization, dimensional reduction, word embedding


lecture_16