CSCI 4150: Introduction to
Articial Intelligence
Thomas P. Trappenberg
Dalhousie University
2009
Acknowledgements
These lecture notes are mostly based on lecture nodes by Andrew Ng from
Stanford University. Specically, we follow basically his lecture note
Learning machines and the
perceptron
This chapter gives a brief historical introduction to learning machines and
neural networks, before treating this area in a more modern fashion in the
following chapters.
5
5.1 Learning Machines
23
5.2 The McCulloch-Pi
1
Introduction and History
A precise denition of articial intelligence (AI) is not easy and often controversial. This introductory chapter outlines some areas associated with AI in a
historical context and by highlighting some modern approaches in AI. Man
Generative models and Naive
Bayes
9.1
CS229 Lecture notes by Andrew Ng, Part
IV
9
CS229 Lecture notes
Andrew Ng
Part IV
Generative Learning algorithms
So far, weve mainly been talking about learning algorithms that model
p(y |x; ), the conditional distrib
Constraint satisfaction
problem (CSP)
4.1
CS221 Lecture notes by Andrew Ng, No. 4
4
CS221 Lecture notes #4
Constraint satisfaction
problems (CSP)
In the previous set of notes, we discussed the application of local search
algorithms to problems such TSP, 8
Learning Theory
8.1
CS221 Lecture notes by Andrew Ng, No 7
see also CS229 Lecture notes by Andrew Ng, Part VI
8
CS221 Lecture notes #7
Supervised learning summary
In the previous sets of notes on supervised learning, we discussed many specic algorithms fo
Articial Intelligence: Heuristic search
Thomas Trappenberg
September 17, 2009
Based on the slides provided by Russell and Norvig, Chapter 4, Section 12,(4)
Romania with step costs in km
71
75
Oradea
Neamt
Zerind
Arad
87
151
Iasi
140
Sibiu
118
80
Timisoara
Unsupervised learning
10.1
K-maens clustering
CS229 Lecture notes by Andrew Ng
10.2
Mixture of Gaussians and EM algorithm
CS229 Lecture notes by Andrew Ng
10.3
The Boltzmann and Helmholts machines
10
CS229 Lecture notes
Andrew Ng
The k -means clustering a
Support vector machines
7.1
CS229 Lecture notes by Andrew Ng, Part
V
7
CS229 Lecture notes
Andrew Ng
Part V
Support Vector Machines
This set of notes presents the Support Vector Machine (SVM) learning algorithm. SVMs are among the best (and many believe i
Part I
Search
Robotics and motion planning
2
To build the controller for a rational agent (robot), we need to nd a way to
describe the problem in a suitable way. In other word, we need to learn how
to translate a real world problem into a description that
Regression, classication and
maximum likelihood
6.1
CS221 Lecture notes by Andrew Ng, No.5
6
CS221 Lecture notes #5
Supervised learning
So far in this course, we have only considered problems where the entire state
of the world is known in advance. Rarely
Search
This chapter is on search and is divided into three parts
(1) Uninformed search (tree search, graph search, etc)
(2) Heuristic search (A*, etc)
(3) Optimization search algorithms (gradient decent, GA, etc)
3.1
CS221 Lecture notes by Andrew Ng, No.