CS 5785 Applied Machine Learning Lec. 7
Prof. Serge Belongie, Cornell Tech
Scribe team: Mor Cohen, Sagar C, Asaf P, Brunno A
Jacqueline P, Kiyan R, Chao S
Sept. 15, 2016
1
Fishers Linear Discriminant
We now continue the discussion on linear discriminants
CS 5785 Applied Machine Learning Lec. 9
Prof. Serge Belongie, Cornell Tech
Scribe: Teja S, Yiqing L, Roger C
Sept. 22, 2016
1
Kernel Density Estimation
During this course we use the term kernel in two separate ways:
1. A device for local weighted averagin
CS 5785 Applied Machine Learning Lec. 10
Prof. Serge Belongie, Cornell Tech
Scribes: Moises Baly, Charlie LaBarge
Sept. 27, 2016
1
Naive Bayes Review
Well continue our discussion on Naive Bayes by working through some examples
from Kevin Murphy: http:/www
CS5785:ModernAnalytics
Prelim
Instructor:Prof.SergeBelongie
11:40am12:55pm,March14,2013
On this exam you are allowed to use a calculator and one 8.5" by 11"sheetofnotes. Thetotal
number of points possible is 30. Writeyouranswers inabluebook oronseparatebl
CS 5785 Applied Machine Learning Lec. 4
Prof. Serge Belongie, Cornell Tech
Scribe: Mor Cohen, Edward W., Angel W., Chloe E., Harsha V.
August, 2016
1
Newton-Raphson
As described at the end of last lecture, to maximize the log likelihood `() we
will make u
CS 5785 Applied Machine Learning Lec. 1
Prof. Serge Belongie, Cornell Tech
Scribe: Mor Cohen, Claire O., Bill M., Asheesh A., Omry S.
August 23, 2016
1
Logistics
Welcome!
Introduce TAs: Yin Cui (office hours on website). Michael Wilber (office
hours on
CS 5785 Applied Machine Learning Lec. 6
Prof. Serge Belongie, Cornell Tech
Scribe: Todd Kawakita, Bochen W., Xinxi C., Liudan X., Sam R.
Sept. 15, 2015
1
Multivariate Gaussian/Normal Density
One of the most commonly used pdfs in engineering is the multiva
CS 5785 Applied Machine Learning Lec. 11
Prof. Serge Belongie, Cornell Tech
Scribe: Fang-Yu (Betty) Chou, Jessie Kuo
Sep. 29, 2016
1
1.1
Cluster Analysis
Review from Last Time
Recall that clustering is an unsupervised method. Central approaches are
protot
CS5785 Homework 0
O VERVIEW
Welcome to CS 5785! After completing this homework, you should be able to find a teammate, set up
your preferred development environment, download and parse a dataset, and use visualization tools to
help you understand what tha
CS5785 Homework 2
The homework is generally split into programming exercises and written exercises.
This homework is due on September 28, 2016 at 11:59 PM EST. Upload your homework to CMS.
Please upload code as a single .zip file and the writeup as a sing
CS 5785 Applied Machine Learning Lec. 8
Prof. Serge Belongie, Cornell Tech
Scribe: Jocelyn Kong, Jeff Ponnor, Harrison Gregg, Ashwin Ramanathan
Sept. 20, 2016
1
Singular Value Decomposition
1.1
Review from Last Time
Recall that the SVD decomposes a rectan
CS5785 Homework 1
The homework is generally split into programming exercises and written exercises.
This homework is due on September 14, 2016 at 11:59 PM EST. Upload your homework to CMS.
Please upload code as a single .zip file and the writeup as a sing
CS 5785 Applied Machine Learning Lec. 2
Prof. Serge Belongie, Cornell NYC Tech
Scribe: Mor Cohen, Nicolas J., Praveen G., Daniel S., Shreyas K, Francesco P.
August 25, 2016
1
Linear Regression
Last lecture we learned about kNN, which is a lazy learning or
Effective Modern C+
Topics include:
The pros and cons of braced initialization, noexcept
specifications, perfect forwarding, and smart pointer make
functions
The relationships among std:move, std:forward, rvalue
references, and universal references
Techni