Machine Learning Week 3
Date: 27th, 29th January, 2004
Scribe: Rajashree Paul, Wendy-Yang Wang, Ming Zhou
Outline
Entropy
Information Gain and Gain Ratio
Selection of the best predictive attribute
Hypothesis Space in Decision Tree Induction
Inductive Bias

Introduction to Computational Game Theory
CMPT 882
Simon Fraser University
Oliver Schulte
Matrix Games and Nash Equilibrium
Matrix Games: Definition
We have n players 1,2,n.
Each player i has a strategy set Si. Members of Si are denoted
by si or ai.
Ea

CMPT - 882
Scribe: March 2/4, 2004
Machine Learning Scribe 7
March 2/4, 2004
Yinan Zhang, Hongyin Cui
Content
Bayes Net
o Inferring Causal Structure (Continued) - Rules for Orientation
Network Traffic Modeling (presented by Leo Chen)
o Bayesian Net (causa

Introduction to Computational Game Theory
CMPT 882
Simon Fraser University
Oliver Schulte
Iterated Weak Dominance
The IWD procedure
In our discussion of trembling-hand equilibrium, we
encountered the idea that players should avoid weakly
dominated strateg

CMPT-882 Scribe Notes on ILP and Inverted Deduction
Scribe Notes on
FOIL and Inverted Deduction
by Gabor Melli ([email protected])
This document summarizes the two inductive logic programming (ILP) approaches of the FOIL
algorithm and induction by inverted ded

Nave Bayes Classifiers
Example: PlayTennis (6.9.1)
Given a new instance, e.g. (Outlook = sunny, Temperature = cool, Humidity =
high, Wind = strong), we want to compute the most likely hypothesis:
n
v NB = arg max P(v j ) P(ai | v j )
vj V
i =1
P(yes)*P(su

Scribe to lecture Tuesday March 16 2004
Scribe outlines:
Message
Confidence intervals
Central limit theorem
Em-algorithm
Bayesian versus classical statistic
Note: There is no scribe for the beginning of the lecture Thursday March 19. The missing part of
t

CMPT 882 Machine Learning
Lecture Notes
Instructor: Dr. Oliver Schulte
Scribe: Qidan Cheng and Yan Long
Mar. 9, 2004 and Mar. 11, 2004
-1-
Basic Definitions and Facts from Statistics
1. The Binomial Distribution
Given a worn and bent coin, there are two p

Introduction to Computational Game Theory
CMPT 882
Simon Fraser University
Oliver Schulte
Decision Making Under Uncertainty
Outline
Choice Under Uncertainty: Formal Model
Choice Principles
o Expected Utility
o Dominance
o Maximin
o Regret
Choice and Unc

Introduction to Computational Game Theory
CMPT 882
Simon Fraser University
Oliver Schulte
Rational Preferences
(start with powerpoint)
Weak Preferences
Let O be a set of options among which an agent A is
choosing. The options can be stocks, ice cream fla

Relationship between Least Squares Approximation and
Maximum Likelihood Hypotheses
Steven Bergner Chris Demwell
,
Lecture notes for Cmpt 882 Machine Learning
February 19, 2004
Abstract
In these notes, a derivation of the estimate of the mean of a normal d

CMPT 882
Computational Game Theory
Simon Fraser University
Instructor: Oliver Schulte
Strategic Analysis of Auctions: Intro
Outline
Common Auction Types
Strategic Analysis of Private Value Auctions: English,
Vickrey, Dutch.
Late Bidding in Internet Auc

CMPT 882
Simon Fraser University
Oliver Schulte
Trade-offs and Social Choice: Pareto-Optimality
Choice Without Uncertainty: Trade-offs
In a choice situation with no uncertainty, the consequences
of each option are known. It may seem that in that case
cho

Scribe Notes from Tuesday January 20
and Thursday January 22, 2004
By:
Maryam Bavarian
Ladan Mahabadi
Irina Pekerskaya
On Tuesday January 20, 2004, a few criticisms of the FIND-S algorithm were discussed:
1) Lack of Convergence of the learner to all possi

Reinforcement Learning
Lecture Notes
Machine Learning, CMPT 882
Instructor: Dr. Oliver Schulte
Spring 2004
Scribe: Mohamed Soliman
March 18th, 23rd, and 25th, 2004
1
Contents
Introduction
The Backgammon World
Example: TD-GAMMON
Control Learning, in Genera

CMPT 882: Machine Learning
spring semester 2004
Scribe of March, 23rd 2004, by Jakob von Recklinghausen ([email protected])
Reinforcement Learning
"Reinforcement learning (.) is a computational approach to learning whereby an
agent tries to maximize the tot

Computational Game Theory
The Basic Denitions
Oliver Schulte
Simon Fraser University
School of Computing Science
February 3, 2010
1
Relations and Functions
Let A be a set. An ordered pair from A is a pair (a, b) such that both a and b
are from A. To say t