Last Updated: 2014 December 1
On Learning and studying
What students should know, but don't
On learning
How to study
How not to study for exams
What one person from the real world wrote about learning
Useful books
- How we Learn: The Surprising Truth About When, Where, and Why it Happens,
Benedict Carey, Random House, 2014.
- Probably Approximately Correct: Nature's Algorithms for learning and
Prospering in a Complex World, Leslie Valiant, Basic Books, 2013.
- Risk Assessment and Decision Analysis with Bayesian Networks,
Norman Fenton, Martin Neil, CRC Press, 2013, p43
About course exams
Exam questions can be based on the following sources: (1) the textbook(s),
(2) readings, (3) lectures, (4) reports, (5) exercises, and (6) on-line notes
and slides. They are based on topics from the beginning of the year up to, and
including, the class before the exam.
Consider all concepts and terminology used in the text book, reports, slides
and lectures and ask the typical questions - how, why, when, where and what -
individually and in combination. In particular, variations are based on
"describe", "explain", "define", "what is meant by", etc.
Some programming exercises in the example programming exercises have been and
may be asked as exam questions.
- Exam instructions for class exams
and the final exam. It is useful to become familiar with them.
- Prolog description questions
- Prolog basic questions
- Prolog accumulator questions
- Prolog cut and not questions
- Prolog chat questions
- Prolog operator questions
- Prolog logic questions
- Search questions
- Expert and knowledge base questions
- Bayesian questions
Example programs
IDA* with instrumentation write statements
Prolog programs from the book Prolog Programming for Artificial
Intelligence, Fourth Edition, by Ivan Bratko, as prepared by the author.
Prolog programs
provided by the course instructor contain the following files.
- binaryTree.pl — an example definition of a binary tree
- chatBasis.pl — the support environment used to produce programs such
as in chatExample.pl.
- chatExample.pl — an example program demonstrating how English
sentences can be used to construct Prolog queries.
- cuttest.pl — examples that show the effects of the cut operator '!'.
- engFreDictionary.pl — example predictes showing how you can define
prefix, postfix and infix operators in Prolog.
- examplePlUnitTesting.pl — an example that shows how, within one file,
Prolog predicates can be defined, and tested using PlUnit.
- listUtilities.pl — Gives definitions of some predicates defining
utility operations on lists.
- maze.pl — example predicates showing show a graph search program can
be developed and various problems that can occur.
- operator.pl — operator examples.
- ring.pl — example predicates based on Wagner's ring cycle operas that
you can use to experiment with.
- sendMoreMoney.pl — Crypt arithmetic problem done with Basic
Prolog and using CLP. Multiple variations show that checking constraints
at appropriate points in a constrained logic program can significantly
increase the efficiency of an algorithm.
- set_print_options.pl — a predicate that changes the amount of a list
that will be displayed.
- sortingAlgorithms.pl — A selection of sort algorithms: bubble, binary
tree, insert, selection, merge, quicksort.
- tower.pl — example predicates showing how compound structures are used
to build data structures.
- zebra.pl — The traditional "Who owns the zebra" logic problem written
in two different ways.
Bayesian Network Examples provided by the course instructor contain
the following files.
- burglarAlarm_fig_14_4.pl — Burglar alarm model Fig 16.4 in
Bratko's book'
- burglarAlarm_multistate.pl — Burglar alarm model using yes-no states
- child_food_choices.pl — a multistate model computing how much
food to buy depending upon children's statistical preferences
- f16_4_BayseianNet.pl — the program in Figure 16.4
- harvardMedical.pl — the Harvard medical model for disease & test
- ModelsForBayesianExamples.pdf — the Bayesian models for the
examples in this package
- normanLate.pl — the Norman late model from the book
Risk Assessment & Decision Analysis with Bayesian Networks
- normanMartinLate.pl — the Norman and Martin late model from the book
Risk Assessment & Decision Analysis with Bayesian Networks
- prosecutorsFallacy.pl — a prosecutor's fallacy model
- terroristAttack.pl — a terrorist attack model
Example report style
An annotated version of the
Sequence ADT describing the structure of a stand-alone report.