U N I V E R S I T Y OF T O R O N T O
Faculty of Arts and Science
A P R I L 2016 EXAMINATIONS
CSC384H1S
Introduction to Artificial Intelligence
Instructors: Sonya Allin and Sheila Mcllraith
Duration 3 hours
Examination Aids: No aids p e r m i t t e d .
You
Note that in this exercise we perform GAC as preprocessing
in advance of any search, as well as during search.
Constraint propagation (GAC, FC, etc.) is often performed in
advance of initiating search as a preprocessing step to prune
variable domains. Onc
KR Tutorial
CSC384 Winter 2016
README - Syntax
Logical implication is commonly included in the syntax of
first-order and propositional logical languages. The symbol used to
denote logical implication differs from language to language. As
such, in any part
UNIVERSITY OF TORONTO
Faculty of Arts and Science
DECEMBER 2014 EXAMINATIONS
Duration 3 hours
No Aids Allowed
CSC384H1F
Introduction to Articial Intelligence
Instructor: Erin Delisle
Student Number:
Last Name:
First Name:
Reminder: A grade of 40%
2/7/17
Generalizing Our Search Problems
Game Tree Search
Chapter 5.1, 5.2, 5.3, 5.6 cover some of the material we cover here.
Section 5.6 has an interesting overview of State-of-the-Art game
playing programs.
Section 5.5 extends the ideas to games with u
Given name:._ Last name:
Student number:_ Signature:
UNIVERSITY OF TORONTO MISSISSAUGA
APRIL 2015 FINAL EXAMINATION
CSC384HES (Introduction to Articial Intelligence)
Instructor: Jing Yan
Duration: 180 minutes
Aids: Non-programmable Calculator Allowed
The
U N I V E R S I T Y OF T O R O N T O
Faculty of Arts and Science
D e c 2015 Examinations
H1F
C S C 384 (Introduction to Artificial Intelligence)
1
Instructor: Fahiem Bacchus
Duration: 3 Hours
E x a m i n a t i o n Aids: A non-programmable calculator is al
_ Fltvaine WWII,
SIGEALURE I _
UNIVERSITY OF TORONTO MISSISSAUGA
APRlL 2016 FINAL EXAMINATION
CSC384H58
Introduction to Artificial Intelligence
Abbas Attarwaia
Duration - 3 hours
Aids: NonProgrammable Calculators
The University of Toronto Mississauga and
UNIVERSITY OF TORONTO
Faculty of Arts and Science
April 2015 Examinations
CSC 384 (Introduction to Articial Intelligence)
Instructor: Fahiem Bacchus
Duration: 3 Hours
Examination Aids: A non-programmable calculator is allowed.
You must obtain a mark of at
Planning
This material is covered in R&N 3rd edition chapter 10
This material is covered in R&N 2nd edition chapters
10.3, 11.1, 11.2, 11.4
Sheila McIlraith, CSC384, University of Toronto, Winter 2016
1
plan n.
1. A scheme, program, or method
worked out b
CSC348: Introduction to Artificial Intelligence
Constraint Satisfaction Problems
(Backtracking Search)
Chapter 6 (R&N, 3rd edition)
6.1: Formalism
6.2: Constraint Propagation
6.3: Backtracking Search for CSP
6.4 is about local search which is a very
Knowledge Representation (KR)
This material is covered in chapters 710 (R&N, 2nd ed) and chapters
7 9 and 12 (R&N, 3rd ed).
Chapter 7 provides useful motivation for logic, and an introduction to
some basic ideas. It also introduces propositional logic, wh
Reasoning under Uncertainty
This material is covered in chapters 13 and 14 of Russell and
Norvig 2nd and 3rd edition.
Chapter 13 gives some basic background on probability from
the point of view of AI.
Chapter 14 talks about Bayesian Networks, which wil
Welcome to CSC384: Intro to Artificial Intelligence
!@#!, MAN.
Sheila McIlraith, University of Toronto, Winter 2016
CSC384: Intro to Artificial Intelligence
Winter 2016
Instructors: Prof. Sheila McIlraith & Dr. Sonya Allin*
Lectures/Tutorials:
Mon/Wed/Fr
Heuristic Search
In uninformed search, we dont try to evaluate which
of the nodes on the frontier/OPEN are most promising.
We never look-ahead to the goal.
Heuristic Search
(Informed Search)
E.g., in uniform cost search we always expand the cheap
1
Solutions to Bayesian Network Problems
1. Given the Bayesian Network about, determine:
(a) if P1 and P5 are independent of P6 given P8
FALSE, the path through P3, P4 and P7 is not blocked; neither P1 and P6 or P5 and P6 are
d-separated.
(b) if P2 is ind
Figure 1: Search Tree for Minimax Problem. MAX nodes are nodes for the player that is seeking
to maximize the value of play, while MIN nodes are nodes for the player seeking to minimize this
value.
1
More Search Probelms
1. Is A s search behavior necessar
Search Problems, University of Toronto, CSC384 - Intro to AI, Fall 2016
1
1: List the order in which nodes are visited in the tree in the figure above for each of the following three
search strategies (choosing leftmost branches first in all cases):
Dept
Acknowledgements
CSC384: Introduction to Artificial Intelligence
These CSC384 slides have been shared and updated by a number of people
including (but not limited to):
Sheila McIlraith
Fahiem Bacchus
Sonya Allin
Craig Boutilier
Hojjat Ghaderi
Rich
Learning Objectives
At the end of the class you should be able to:
define a directed graph
represent a problem as a state-space graph
explain how a generic searching algorithm works
c
D.
Poole and A. Mackworth 2010
Artificial Intelligence, Lecture 3.1, Pa
unvaxvum yum-.7.- Av . w.J ._._.-_._ _
goal statesfor example, the two dirtfree goal states in Figure 3. 3then we can construct a
new dummy goal state whose immediate predecessors are all the actual goal states. But if the-
goal 1s an abstract description
1
More Bayesian Network Problems
1. Given the Bayesian Network about, determine:
(a)
(b)
(c)
(d)
if
if
if
if
P1
P2
P1
P1
and P5 are independent of P6 given P8
is independent of P6 given no information
is independent of P2 given P8
is independent of P2 and
Sample Questions
(Search)
1
Short Answer
1. Is A s search behavior necessarily exponentially explosive?. That is, does its search time
always grow at least exponentially with the length of the optimal solution.
2. It would seem that iterative deepening se
CSC384 Bayes Net Sample Questions
Fall 2016
1
Bayes Nets
1. Two astronomers in dierent parts of the world make measurements M1 and M2 of the number
of stars N in some small region of the sky, using their telescopes. Normally, there is a small
probability
Forward Checking Algorithm
Forward Checking Algorithm
FC(Level) /*Forward Checking Algorithm */
If all variables are assigned
PRINT Value of each Variable
RETURN or EXIT (RETURN for more solutions)
(EXIT for only one solution)
V := PickAnUnassignedVariabl
RECALL: Problem Formulation
To formulate a problem as a search problem we need the
following components:
Modeling Problems
and
Constructing Heuristics
Sheila McIlraith, CSC384, University of Toronto, Winter 2016
1. Formulate a state space over which to se
Learning Objectives
At the end of the class you should be able to:
devise an useful heuristic function for a problem
demonstrate how best-first and A search will work on a graph
predict the space and time requirements for best-first and A
search
c
D.
Pool
Learning Objectives
At the end of the class you should be able to:
demonstrate how depth-first search will work on a graph
demonstrate how breadth-first search will work on a graph
predict the space and time requirements for depth-first and
breadth-first