CS 486/686  Assignment #1 Sample solution
Due Date: Wednesday, June 4 by 5pm.
Note: The following exercises are to be done individually. No late assignments will be accepted. Please
hand in your assignment using the drop boxes on the 4th floor of MC by 5
Applications and
Ethical Issues
CS 486/686:Introduction to Articial Intelligence
Fall 2013
1
Example Applications
Security Games

Limited security resources: Selective
checking
Adversary monitors defenses, exploits
patterns
How do you assign limited reso
Machine Learning
[RN2] Sec 18.118.3
[RN3] Sec 18.118.3
CS 486/686
University of Waterloo
Lecture 15: March 4, 2014
Outline
Inductive learning
Decision trees
CS486/686 Lecture Slides (c) 2014 P. Poupart
2
What is Machine Learning?
Definition:
A compu
Statistical Learning
[RN2 Sec 20.120.2]
[RN3 Sec 20.120.2]
CS 486/686
University of Waterloo
Lecture 17: March 11, 2014
Outline
Statistical learning
Bayesian learning
Maximum a posteriori
Maximum likelihood
Learning from complete Data
CS486/686 Lec
Statistical Learning (II)
[RN2] Sec 20.3
[RN3] Sec 20.3
CS 486/686
University of Waterloo
Lecture 18: March 13, 2014
Outline
Learning from incomplete Data
EM algorithm
CS486/686 Lecture Slides (c) 2014 P. Poupart
2
Incomplete data
So far
Values of all
Course wrap up
CS 486/686
University of Waterloo
Lecture 24: April 3, 2014
Outline
Course wrap up
Final exam info (see course website)
Other AI courses
AI research
AI jobs
CS486/686 Lecture Slides (c) 2014 P. Poupart
2
Topics Covered
Search algorithms
Pro
Bandits
CS 486 / 686
University of Waterloo
Lecture 22: March 27, 2014
Exploration/Exploitation Tradeoff
Fundamental problem of RL due to the active
nature of the learning process
Consider onestate RL problems known as
bandits
2
CS486/686 Lecture Slide
Homework Assignment 1: Search Algorithms
CS486/686 Spring 2015
Instructor: Pascal Poupart
Out: May 12, 2015
Due: May 27 (11:59 pm), 2015. Submit an electronic copy of your assignment via LEARN. Late
assignments may be submitted within 24 hrs for 50% credi
Machine Learning
CS 486/686: Introduction to Articial Intelligence
Fall 2013
1
Outline
Forms of Learning
Inductive Learning

Decision Trees
2
What is Machine Learning
Denition:

A computer program is said to learn from
experience E with respect to so
CS 486/686: Introduction to Articial Intelligence
FOL Example
Knowledge Base in a Natural Language
1. Marcus is a person.
2. Marcus is a Pompeian.
3. All Pompeians are Roman.
4. Caesar is a ruler.
5. All Romans are either loyal to Caesar or hate Caesar.
6
Constraint Satisfaction
[RN2] Sec 5.15.2
[RN3] Sec 6.16.3
CS 486/686
Lecture 4: January 16, 2014
University of Waterloo
1
CS486/686 Lecture Slides (c) 2014 P. Poupart
Outline
What are CSPs?
Standard search and CSPs
Improvements
Backtracking
Backtra
Probabilistic Reasoning
[RN2] Sections 14.1, 14.2
[RN3] Sections 14.1, 14.2
University of Waterloo
CS 486/686
Lecture 7: January 28, 2014
Outline
Review probabilistic inference,
independence and conditional
independence
Bayesian networks
What are they
Bayes Nets (continued)
[RN2] Section 14.4
[RN3] Section 14.4
CS 486/686
University of Waterloo
Lecture 8: January 30, 2014
Outline
Inference in Bayes Nets
Variable Elimination
2
CS486/686 Lecture Slides (c) 2014 P. Poupart
Inference in Bayes Nets
The i
Uncertainty
[RN2 Sec. 13.113.6]
[RN3 Sec. 13.113.5]
CS 486/686
University of Waterloo
Lecture 6: January 23, 2014
CS486/686 Lecture Slides (c) 2014 P. Poupart
1
A Decision Making Scenario
You are considering to buy a used car
Is it in good condition?
H
Reasoning Over Time
[RN2] Sec 15.115.3, 15.5
[RN3] Sec 15.115.3, 15.5
CS 486/686
University of Waterloo
Lecture 13: February 24, 2014
Outline
Reasoning under uncertainty over time
Hidden Markov Models
Dynamic Bayesian Networks
2
CS486/686 Lecture Sli
Decision Networks
[RN2] Sections 16.5, 16.6
[RN3] Sections 16.5, 16.6
CS 486/686
University of Waterloo
Lecture 12: February 13, 2014
Outline
Decision Networks
Aka Influence diagrams
Value of information
2
CS486/686 Lecture Slides (c) 2014 P. Poupart
D
Utility Theory
[RN2] Sect 16.116.3
[RN3] Sect 16.116.3
CS 486/686
University of Waterloo
Lecture 11: February 11, 2014
1
CS486/686 Lecture Slides (c) 2014 P.Poupart
Outline
Decision making
Utility Theory
Decision Trees
Chapter 16 in R&N
Note: Some
Assignment 1 Solution and Marking Scheme
1. Informed Search (40 points)
1(a) (6 points: 2 for the correct answer and 4 for the proof)
Consider H1 = Misplaced Tile Heuristic and H2 = Manhattan Distance Heuristic.
Acceptable proofs should have that H2 is be
Assignment 2: Bayesian Networks and Decision Networks
CS486/686 Spring 2015
Out: May 28, 2015
Due: June 12 (11:59 pm), 2015. Submit an electronic copy of your assignment via LEARN. Late
assignments may be submitted within 24 hrs for 50% credit.
Be sure to
CS 486/686 Spring 2015
Assignment 2 Solutions
June 15, 2015
Question 2a
1
Question 2b
Part I  Prior Probability
P r(f raud) = P r(f raudtrav)P r(trav) + P r(f raudtrav)P r(trav)
= 0.01 0.05 + 0.004 0.95
= 0.0043
Part II  Posterior Probability
OC
t
f
f