Stats 315B Homework 1 Solutions
Problem 1
First build a large tree so we can use rparts cross-validation feature to determine the optimal tree size. The cross-validation errors can be found using
the function printcp and used to nd the optimal value of th
Statistics 315a
Homework 1, due Wednesday January 28, 2015.
ESL refers to the course textbook, and ESL 2.4 refers to exercise 2.4 in
ESL. Since the homework assignments count 70% of your nal grade, you
must do them on your own. Problem 1 is computing inte
Stats315B Homework 2
Spring 2015
Due: May 21, 2015
Please keep answers short. Verbosity will not be rewarded.
1. Random forests predict with an ensemble of bagged trees each trained on a bootstrap
sample randomly drawn from the original training data. Add
ESL Chapter 3 Linear Methods for Regression
Trevor Hastie and Rob Tibshirani
Linear Methods for Regression
Outline
The simple linear regression model
Multiple linear regression
Model selection and shrinkagethe state of the art
1
ESL Chapter 3 Linear Me
Stat315B
Methods for Applied Statistics
Homework 3
Due date: Friday, June 5, 2015.
1. Consider a multi-hidden layer neural network trained by sequential steepest-descent using the
weight updating formula
wt = wt1 G(wt1 ).
Here t labels the observations pr
SLDM III c Hastie & Tibshirani - March 18, 2010
Dimension Reduction and SVD
Principal Components
Suppose we have N measurements on each of p variables
Xj , j = 1, . . . , p. There are several equivalent approaches to
principal components:
Produce a deriv
ESL Chapter 2 Overview of Supervised Learning
Trevor Hastie and Rob Tibshirani
Overview of Supervised Learning
Notation
X: inputs, feature vector, predictors, independent variables.
Generally X will be a vector of p real values. Qualitative features are
Stats315a: Statistical Learning
1
Statistics in the news
How IBM built Watson, its Jeopardy-playing
supercomputer by Dawn Kawamoto DailyFinance 02/08/2011
Learning from its mistakes
According to David Ferrucci
(PI of Watson DeepQA technology for IBM Resea