Can someone explain how numeric values were derived for P(fringe)? I
see in the book where it states, "Notice that the expression
P(b|known,P[1,3],fringe) is 1 when fringe is consistent with the
breeze observations and 0 otherwise".  If I wanted to determine the
following:

P(b,known,P[1,3]|fringe)

I would need to know P(fringe).

I've read in the 2nd edition, that a node is conditionally independent
of its non-descendants given its parents or given its Markov blanket.
The 2nd edition makes reference to the 1st edition regarding d-
separation and I have read the 1st edition's information about d-
separation; however, I have come across in some additonal reading
something called Berkson's paradox.

This paradox seems to show circumstances that contradict the
statements made about d-separation and Markov blankets.  Am I correct
in thinking that Berkson's paradox is an exception to rules regarding
d-separation and Markov blankets?

On Thu, May 03, 2007 07:37:59PM -0000, orondojones wrote:
> I've read in the 2nd edition, that a node is conditionally independent
> of its non-descendants given its parents or given its Markov blanket.
> The 2nd edition makes reference to the 1st edition regarding d-
> separation and I have read the 1st edition's information about d-
> separation; however, I have come across in some additonal reading
> something called Berkson's paradox.
>
> This paradox seems to show circumstances that contradict the
> statements made about d-separation and Markov blankets.  Am I correct
> in thinking that Berkson's paradox is an exception to rules regarding
> d-separation and Markov blankets?

The so called Berkson's paradox states:

         if 0 < P(A) < 1 and 0 < P(B) < 1,
         and P(A|B) = P(A), i.e. they are independent,
         then P(A|B,C) < P(A|C) where C = A∪B (i.e. A or B).

Where is the contradiction?

--
Iván F. Villanueva B.

Let it first be said that I applaud Russell and Norvig for including
section 26.3 in the book, regardless of my comments thereupon.

*

p. 963:  "The "intelligence explosion" has also been called the
technological singularity by mathematics professor and science fiction
author Vernor Vinge, who writes (1993), "Within thirty years, we will
have the technological means to create superhuman intelligence.
Shortly after, the human era will be ended."

Vernor Vinge coined the term "Singularity" after trying and failing to
write a science fiction story in which the hero was smarter than
human.  Vinge found that he could not write the story, because he was
unable to realistically envision such a hero.  As Vinge later wrote
(_True Names and Other Dangers_ short story collection, p. 47):

"Here I had tried a straightforward extrapolation of technology, and
found myself precipitated over an abyss.  It's a problem we face every
time we consider the creation of intelligences greater than our own.
When this happens, human history will have reached a kind of
singularity - a place where extrapolation breaks down and new models
must be applied - and the world will pass beyond our understanding."

Vinge originally coined "Singularity", not by analogy to a
"singularity" of a function, but by analogy to the "singularity" at
the center of a black hole, where (circa 1983) physicists' models
broke down and gave nonsensical answers.  The original "Singularity"
was purely epistemological - it was a breakdown in your *model of*
reality, not reality itself.  It's pretty hard to extrapolate your
model of the future (said Vinge) past the point where that model
starts predicting the existence of minds substantially smarter than human.

As far as I know this is Vinge's original observation, and it is
logically distinct from I. J. Good's earlier thesis of the
intelligence explosion.  You'd have to separately argue whether or not
an AI could recursively self-improve; versus whether or not a
smarter-than-human AI would produce a substantial discontinuity with
today's world (I would argue that it depends on the initial conditions
of the AI).  You can imagine A without B, B without A, neither, or both.

Especially with regard to the "Singularity", which seems rapidly to be
turning into a "suitcase word" a la Minsky, it is critical to keep
distinct theses distinct.  Otherwise someone may define the thesis one
way, argue for another thesis, and present a third thesis as having
been proven.  Today Vinge's original thesis is sometimes referred to
as the "Event Horizon" to distinguish it from other, later usages of
the word "Singularity".

*

p. 963:  "Good and Vinge (and many others) correctly note that the
curve of technological progress is growing exponentially at present
(consider Moore's Law).  However, it is quite a step to extrapolate
that the curve will continue on to a singularity of near-infinite growth."

Moore's Law is logically distinct from both I. J. Good's intelligence
explosion and from Vinge's event horizon.  If anything, Vinge's thesis
argues against the indefinite continuation of Moore's Law, because it
would be surprising if industry models developed for human
corporations worked for predicting events after the advent of
smarter-than-human AIs.

Vinge's thesis is about an absolute threshold which could be
approached by exponential progress, linear progress, logarithmic
progress, etc.  As far as I know, Moore's Law did not become
associated with Vinge's Singularity until Vinge started trying to
predict the time of his Singularity - a whole different ballpark from
the core thesis itself!  You need far more knowledge, and not just
about aerodynamics, to say "Someone will build a flying machine in
1905" instead of "Someone will build a flying machine eventually."  A
date might not be predictable even in principle.  If you took the
world from 1880 and reran the planet, the first flying machine might
be built ten years earlier or later.  It may not even be useful to
think of an absolute level of processing power as "necessary for AI".
   Rather, the more processing power the researchers have, the less
clever they can be and still build AI.  Even this is not always true;
if the one does not understand regularization, then building a neural
network with a billion times as many units just implies a billion
times as much overfitting.

I. J. Good's thesis is that, if you can make yourself a little
smarter, you are then able to see more ways to make yourself even
smarter, and so a chain reaction of self-improvement occurs.  This
chain reaction does not need to continue forever, nor follow an exact
exponential curve, in order to be of great practical significance.
Biological neurons fire at less than 200Hz; biological axons transmit
messages at 150 meters/second which is less than a millionth the speed
of light; and each spike dissipates around a million times as much
energy as the thermodynamic minimum at 300 Kelvin.  The laws of
physics definitely permit the construction of a computer at least a
million times as fast as the human brain without shrinking the brain
or cooling the brain.  Even if the intelligence explosion tops off at
exactly this point, the existence of minds a million times faster than
human (never mind "a million times as smart") would <understatement
size='huge'>be of great practical significance</understatement>.

*

p. 963:  "Ray Kurzweil, in The Age of Spiritual Machines (2000),
predicts that by the year 2099 there will be "a strong trend toward a
merger of human thinking with the world of machine intelligence that
the human species initially created.  There is no longer any clear
distinction between humans and computers."  There is even a new word -
transhumanism - for the active social movement that looks forward to
this future.  Suffice it to say that such issues present a challenge
for most moral theorists, who take the preservation of human life and
the human species to be a good thing."

The FAQ of the World Transhumanist Association defines transhumanism
as "The intellectual and cultural movement that affirms the
possibility and desirability of fundamentally improving the human
condition through applied reason, especially by developing and making
widely available technologies to eliminate aging and to greatly
enhance human intellectual, physical, and psychological capacities."
Cyborgs are not explicitly mentioned, either for inclusion or
exclusion - it's about generalized persons in general.

It is widely agreed that if a young child falls on the train tracks,
there is a moral duty to pull them away.  It is widely agreed that if
someone of age 50 suffers from a debilitating disease that decreases
their quality of life, it is good to cure them.  Now if you have a
logical turn of mind, you are bound to ask whether this is a special
case of a general ethical principle which says "Life is good, death is
bad; health is good, sickness is bad" and, if so, whether it would be
a good thing to extend lifespan and healthspan out to 150 years, not
just 75 years.  Many people feel an instinctive shock at this, because
it is not an accustomed idea, and they will rationalize reasons why
150 years of health is a dangerous and subversive notion, much worse
than 55 years of health followed by 20 years of sickness followed by
death.  But one who reads scientific history and has a sense of
temporal perspective might remember that anesthetics and the smallpox
vaccine were viewed with great suspicion by the
bioethicist-equivalents of that day.

"Transhumanist" ethics are actually simpler - can be specified with
fewer bits - because they are consistent in their judgments; life is
good, death is bad, health is good, sickness is bad, and there is no
special exception to this rule when you extend lifespan and healthspan
beyond 75, or when you use startling new technologies to get them.
Once you see a happy, healthy person, you're done, whether they
previously lived 40 years or 140 years, and whether they're made of
carbon or silicon.

So a transhumanist analytic philosopher (e.g. Nick Bostrom) would not
say that there is anything inherently desirable about cyborging
yourself, but would also say that there is nothing inherently
undesirable about it.  The prospect is interesting insofar as it may
be a useful means to such normative ends as health, vigor, or
intelligence.  And while _Wired_ editors may get a kick (and extra
sales) out of using shocking futuristic phrases like "becoming
indistinguishable from our machines", this doesn't become desirable
(to a transhumanist) just because it sounds shocking and futuristic;
you'd have to make a case under utilitarian ethics.  A transhumanist
would be open to that case - they wouldn't run away, screaming,
"Machines!  Ew!" - but they would still demand that you make the
argument under consequentialism.

*

p. 963:  "For the most part, it seems that robots are the protagonists
of so many conquer-the-world stories because they represent the
unknown, just like the witches and ghosts of tales from earlier eras.
   Do they pose a more credible threat than witches and ghosts?  If
robots are properly designed as agents that adopt their owner's goals,
then they probably do not: robots that derive from incremental
advances over current designs will serve, not conquer.  Humans use
their intelligence in aggressive ways because humans have some
innately aggressive tendencies, due to natural selection.  But the
machines we build need not be innately aggressive, unless we decide to
build them that way."

I agree with the general thrust but its achievement is being taken too
much for granted.  If most technologies' negative repercussions tend
to be substantially outweighed by their positive aspects, that is a
historical generalization which implicitly takes into account the
professional paranoia that scientists and engineers exert to keep
things that way.  A modern nuclear power plant is not safe because the
engineers involved waved off all objections by saying that Technology
Is the March of Human Progress, but because the engineers spent all
night worrying about how the reactor design might fail.  In the same
way that Moore's Law would grind to a halt if all the chip-design
researchers decided to take a vacation, technological consequences
stop being positively skewed when engineers stop being pessimistic.

Indeed, an AI need not share human aggressive tendencies, unless we
design them that way; and an AI need not share human positive
tendencies, unless we understand how to design them that way and
successfully do so.  All human beings have roughly the same cognitive
architecture; we all have a prefrontal cortex, cerebellum, amygdala,
etc.  In the space of all possible mind designs, all human beings are
packed into one small dot.  What the term "Artificial Intelligence"
really refers to is all the rest of mind design space; there are
enormously more possible AIs than possible humans.  To reach into that
gigantic space, and pluck out a highly intelligent agent with a
knowably positive impact on the world relative to our utility
function, is a technical challenge of very high order.

Humanity has solved high-order technical challenges before; it wasn't
exactly easy to walk on the Moon.  Here, the big problem is that it
might be significantly easier to construct a highly intelligent agent
that is *not* carefully shaped to a positive outcome - easier to build
an unFriendly AI than a Friendly AI.  It's hard to see how it could be
otherwise, since the design space of highly intelligent AIs is a
superset of the space of highly intelligent Friendly AIs.  If I. J.
Good is right about the intelligence explosion effect, this could be a
really severe dilemma for humanity - we'd have to solve the harder
technical challenge *first*.

Many current AI techniques, such as gradient descent in neural
networks, produce systems that are intelligent but to a large degree
opaque.  Evolutionary programming produces code that may partially
match an optimization criterion, but does not match it exactly, and
may produce wildly different behaviors in contexts outside the
training problems.  These techniques, which are still growing in
power, are not at all well-suited to shaping a knowably Friendly AI.
"So use Bayesian decision theory," you say, and I agree in principle.
   But there are major design challenges in building a reflective
Bayesian agent that self-modifies and self-improves, but maintains an
invariant optimization target relative to the outside world (a utility
function or generalization thereof).  We do not currently know how to
do this.  It will take new mathematics.  Current decision theory would
go into an infinite loop and crash if you tried to use it to calculate
the expected utility of modifying the part of the AI that does the
self-modifications.  It's a math problem, and I think it's a solvable
math problem, but someone has to actually solve it or there may not be
a positive outcome for humanity.

All this, of course, is a rather long story; see my book chapter
"Artificial Intelligence as a positive and negative factor in global
risk", draft online at http://singinst.org/AIRisk.pdf.  But in
summary, there are specific design challenges in ensuring a positive
outcome - good will is not enough, there must be knowledge - and
success on these challenges should not be taken for granted by the
next generation of AI researchers.

--
Eliezer S. Yudkowsky                          http://singinst.org/
Research Fellow, Singularity Institute for Artificial Intelligence

I'd like to begin my comments on "Artificial Intelligence: A Modern
Approach, 2nd Edition" by saying that this is a truly, truly excellent
book, which I recommend to all technically minded friends when they
say they want to learn more about AI, and whenever someone takes a
first college course on AI I tell them to make sure it is taught with
this book.  There are just a few minor comments and errata...

I've divided this email into two parts, since the comments on page 963
are extensive enough to deserve a separate email.

*

The only actual math error listed in my notes is on page 719.  There's
a stray minus sign in the first equation of (20.4), for the
maximum-likelihood estimate of a gaussian's mean.  Since the
derivative is subsequently set equal to zero, the error does not
propagate, but the qualitative physics is wrong - if all x_j are
higher than u, increasing u should increase the log-likelihood, not
decrease it.

*

On page 9, it says that "Thomas Bayes (1702-1761) proposed a rule for
updating probabilities in the light of new evidence.  Bayes' rule and
the resulting field called Bayesian analysis form the basis of most
modern approaches to uncertain reasoning in AI systems."  IANAHOM (I
Am Not A Historian Of Mathematics) but I've read in more than one
source that Bayes did not invent Bayes's Theorem.  For example, E. T.
Jaynes's _Probability Theory: The Logic of Science_, section 4.6.1 on
page 112:

"...the kind of calculations we are doing are called 'Bayesian'.  We
shall follow this long-established custom, although it is misleading
in several respects.  The general result (4.3) is always called
'Bayes' theorem', although Bayes never wrote it; and it is really
nothing but the product rule of probability theory which had been
recognized by others, such as James Bernoulli and A. de Moivre (1718),
long before the work of Bayes.  Furthermore, it was not Bayes but
Laplace (1774) who first saw the result in generality and showed how
to use it in real problems of inference.  Finally, the calculations we
are doing - the direct application of probability theory as logic -
are more general than mere application of Bayes' theorem; that is only
one of several items in our toolbox."

Bayes actually derived what is now known as Laplace's Rule of
Succession.  Laplace, impressed by this use of inverse inference,
decided to call such calculations "Bayesian".  Thus, Bayesian
probability theory should really be known as "Laplacian probability
theory", Laplace's Rule of Succession should really be known as
"Bayes's Rule of Succession", and I don't think anyone knows who
invented Bayes's Theorem.

*

My highest-priority want item for a third edition would be at least
one dedicated section, with examples, for the field of heuristics and
biases - the box on p. 592 isn't nearly enough.  This is a hugely
important and growing field, critical to anyone interested in taking
apart human cognition to see how it works, and helpful to any human
being who wants an owner's manual for their own reasoning.  It would
be a significant service to students to show them enough of the basics
to tantalize them into learning further.  I would recommend
introducing at least availability, anchoring and adjustment, and the
conjunction fallacy.

(See e.g. http://singinst.org/Biases.pdf).

--
Eliezer S. Yudkowsky                          http://singinst.org/
Research Fellow, Singularity Institute for Artificial Intelligence

On Sat, May 26, 2007 01:26:22PM -0700, Eliezer S. Yudkowsky wrote:
> I'd like to begin my comments on "Artificial Intelligence: A Modern
> Approach, 2nd Edition" by saying that this is a truly, truly excellent
> book, which I recommend to all technically minded friends

But the content of the book is not free. It would be great if the
authors will release at least some parts (e.g. the Summaries and the
Bibliographical and Historical Notes) in a Creative Commons License.
For instance I would like to incorporate some parts in my open source project,
and others maybe interested in doing the same on Wikipedia articles.

--
Iván F. Villanueva B.
A.I. project http://www.artificialidea.com

Hey Ivan and Eliezer,

BTW, I think you folks may also want to voice these opinion on the
other Yahoo Group.

<aima-instructors@yahoogroups.com>

--
Take care,

Muaz Niazi
Asst. Prof.,
Foundation University,
FUIMCS, 1 New Lalazar,
Rawalpindi,
Pakistan


"A common mistake that people make when trying to design something
completely foolproof is to underestimate the ingenuity of complete
fools. "

Douglas Adams, Mostly Harmless
English humorist & science fiction novelist (1952 - 2001)


On 5/27/07, Ivan F. Villanueva B. <ivan@... > wrote:
>
> On Sat, May 26, 2007 01:26:22PM -0700, Eliezer S. Yudkowsky wrote:
>  > I'd like to begin my comments on "Artificial Intelligence: A Modern
>  > Approach, 2nd Edition" by saying that this is a truly, truly excellent
>  > book, which I recommend to all technically minded friends
>
>  But the content of the book is not free. It would be great if the
>  authors will release at least some parts (e.g. the Summaries and the
>  Bibliographical and Historical Notes) in a Creative Commons License.
>  For instance I would like to incorporate some parts in my open source
project,
>  and others maybe interested in doing the same on Wikipedia articles.
>
>  --
>  Iván F. Villanueva B.
>  A.I. project http://www.artificialidea.com
>

hi, I want to find all solutions in state space for a
problem with dfs algorithm I write this I think it is
true but not efficient. Can anyone say something about
it or write a new algorithm!

initial(parameter for method dfs): x = root
initial(global variable): path.add(root);
initial(global variable): dynamicArray solutions;

method dfs(node x, dynamicArray path[])
{
path.add(x)
foreach ( node i in successor(x) )
{
	 if( GoalTest(i)==true )
		 solutions.add(path);
	 else
	 //we must have new copy if we pass just a reference
	 //it is not work correctly!
		 dfs(i,deepCopy(path))

}
return;
}


       ___________________________________________________________
Yahoo! Mail is the world's favourite email. Don't settle for less, sign up for
your free account today
http://uk.rd.yahoo.com/evt=44106/*http://uk.docs.yahoo.com/mail/winter07.html

> On 5/27/07, Ivan F. Villanueva B. <ivan@...> wrote:
>>     But the content of the book is not free. It would be great if the
>>     authors will release at least some parts (e.g. the Summaries and the
>>     Bibliographical and Historical Notes) in a Creative Commons License.

On Sun, May 27, 2007 10:33:10AM -0700, Peter Norvig wrote:
> That's an interesting idea.  Of course the publishers, not the authors, are
> thee ones with the right to do that.  But the authors can encourage the
> publishers to do so.

That's probably because the authors, not the publishers, didn't care enough
about the rights of the published work and the specific parts of the contract.

I would suggest a license like that for as many parts of the book as possible:

     http://creativecommons.org/licenses/by-nc-sa/3.0/

That would be really great. We are in the 21. century, with amazing new things
like Wikipedia. Help the new generation!

--
Iván F. Villanueva B.
A.I. Open Source project:      --     www.artificialidea.com
FFII.org Deutschland           --     de.ffii.org
FFII.org España                --     es.ffii.org

Hello!
I tried out  with calculation of the Normalization constant(alpha') on
page 485, for the Wumpus example, but i failed.
Could you help me in that?
Thanks in advance!

I'm late to the game, some would say.  But I've been interested in the
concepts of AI since a kid.  Only recently have I started doing
something about it.  I'm a PHP engineer and have been coding for 7
years.  Got the architecture and logic side down.

My big issue now is math.  As I learn more and breadth/depth searches,
neural networks, etc I'm realizing I'm really deficient in the math
department.

So I'm trying to figure out a good path for the next year to study.  I
skipped college and got as far as Algebra, Geometry and a hint of Trig
in High School.

Any suggestions on what I should start with and a possible path I
should  follow to help me during my AI learning process?  Thanks!

Hi,

I have some questions regarding chapter 13.7 "The Wumpus World
Revisited" and the derivation of the distribution of P1,3 given known
and breeze evidence in particular:

1.) The probability that a particular cell contains a pit is 0.2 = 3/15,
so that pits are distributed over every cell except for a starting cell,
here 1,1. Now, when calculating the probability of a configuration
containing n pits in Equation 13.15 the product of each cell's
probability distribution is used because of the absolute independence
relationships. The resulting term is 0.2^n x 0.8^(16-n), although the
starting cell is not allowed to contain a pit, shouldn't it be 0.2^n x
0.8^(15-n) x 1 for configurations with no pit in the starting cell and 0
for configurations with a pit in the starting cell because of a
distribution of <0,1> for the starting cell to contain a pit?

2.) P(fringe) is the prior probability of a fringe configuration.
However, when showing the five possible models of the fringe in Figure
13.7 those three possibilities that do not hold the evidence of
perceived breezes in 1,2 and 2,1 are not shown. Does this fact mean that
P(fringe) is in any kind conditioned on the evidence or is it still the
prior distribution with those three combinations not shown?

3.) Later, P(known) is taken out of the summation and put into the
normalizing constant, the sum over the possible configurations of other
is said to be 1. But, that the sum over other of P(other) is 1 is
unclear to me. Of course, having 12 cells in unknown I have 2^12 = 4096
configurations of pits in these cells and a probability of 1/(2^12) to
get one particular configuration. Summing over 2^12 terms results of
course in 1. Same for fringe with 2 cells (2^2 configurations with a
probability of 1/(2^2) each) and other with 10 cells (2^10
configurations with a probability of 1/(2^10) each). But how is this
related to the prior distribution of <0.2,0.8> of each cell and the
product in Eq. 13.15 for the prior of one configuration?

Using combinatorics I have n!/( k!(n-k)! ) combinations to distribute k
holes over n cells, e.g. for distributing 3 pits over 15 cells
15!/(3!x12!) = 455 combinations.
Considering the 5 possible models in Figure 13.7, we have three cases
for other:
3 pits in the fringe: no pit in other - 10!/(0!x10!) * 0.2^0 * 0.8^10
2 pits in the fringe: one pit in other - 10!/(1!x9!) * 0.2^1 * 0.8^9
1 pit in the fringe: two pits in other - 10!/(2!x8!) * 0.2^2 * 0.8^8

However, these values do, of course, not sum up to 1. Even adding the
case, that is not shown in Figure 13.7 with three pits in other does not
correct that, neither does the summing over all 8 possibilities of pits
in the fringe. Actually the above equations do only fulfill our goal of
having the sum of other equal to 1, if we extend those three cases to
distribute every possible number of holes (with k ranging from 0 to 10
and n being 10) over the cells in other. Am I missing something out
here, e.g. a normalization, or am I on a completely misleading way?
So the fundamental question is:
"Is sum_other P(other) = 1?", "Why is it one?"
and, furthermore, is the result of the summation relevant at all or
could it just be put in the normalizing constant just as we did with
P(known)?

Despite my lousy explanations, I hope that my concerns are
understandable and I hope to receive some comprehensible answers.


Best regards
Dirk

Can you elaborate more on what the problem expects us to do in the
vice versa case?

thanks
kanishka

i want to follow this book in my teaching of AI course. when i was
taught this course at that time i worked on Prolog , but now i want to
work on LISP , i have intel machines , will please someone tell me
which LISP version is best for the codes of this book. and can i
download  that for educational purpose.
abdul samad

---

From: aima-talk@yahoogroups.com [mailto:aima-talk@yahoogroups.com] On Behalf Of a_s_khan_b
Sent: Wednesday, September 12, 2007 2:24 PM
To: aima-talk@yahoogroups.com
Subject: [aima-talk] as an instructor i need a bit help

i want to follow this book in my teaching of AI course. when i was
taught this course at that time i worked on Prolog , but now i want to
work on LISP , i have intel machines , will please someone tell me
which LISP version is best for the codes of this book. and can i
download that for educational purpose.
abdul samad

On Wed, Sep 12, 2007 06:24:13PM -0000, a_s_khan_b wrote:
> i want to follow this book in my teaching of AI course. when i was
> taught this course at that time i worked on Prolog , but now i want to
> work on LISP , i have intel machines , will please someone tell me
> which LISP version is best for the codes of this book. and can i
> download  that for educational purpose.
> abdul samad

More about the Lisp Code of AIMA:
http://aima.cs.berkeley.edu/lisp/doc/overview.html

For Intel machines there is of course Linux variants like www.debian.org which
includes GNU CLISP, a Common Lisp implementation. Debian and GNU CLISP are
completely free.

However, I don't know if the Lisp Aima Code works well with GNU CLISP?

A list of other free Lisp implementations is on:

http://en.wikibooks.org/wiki/Programming:Common_Lisp/Installation

--
Iván F. Villanueva B.
http://www.ffii.org

Are there any misleading heuristics apart from the non-admissible ones
that would make A* perform inefficiently? I found one example in this
CI  space Search tool
(http://www.cs.ubc.ca/labs/lci/CIspace/Version4/search/help/tutorial2.html).
  Can anyone tell me why it is misleading?

Hi,

I have some questions regarding chapter 13.7 "The Wumpus World
Revisited" and the derivation of the distribution of P1,3 given known
and breeze evidence in particular:

1.) The probability that a particular cell contains a pit is 0.2 = 3/15,
so that pits are distributed over every cell except for a starting cell,
here 1,1. Now, when calculating the probability of a configuration
containing n pits in Equation 13.15 the product of each cell's
probability distribution is used because of the absolute independence
relationships. The resulting term is 0.2^n x 0.8^(16-n), although the
starting cell is not allowed to contain a pit, shouldn't it be 0.2^n x
0.8^(15-n) x 1 for configurations with no pit in the starting cell and 0
for configurations with a pit in the starting cell because of a
distribution of <0,1> for the starting cell to contain a pit?

2.) P(fringe) is the prior probability of a fringe configuration.
However, when showing the five possible models of the fringe in Figure
13.7 those three possibilities that do not hold the evidence of
perceived breezes in 1,2 and 2,1 are not shown. Does this fact mean that
P(fringe) is in any kind conditioned on the evidence or is it still the
prior distribution with those three combinations not shown?

3.) Later, P(known) is taken out of the summation and put into the
normalizing constant, the sum over the possible configurations of other
is said to be 1. But, that the sum over other of P(other) is 1 is
unclear to me. Of course, having 12 cells in unknown I have 2^12 = 4096
configurations of pits in these cells and a probability of 1/(2^12) to
get one particular configuration. Summing over 2^12 terms results of
course in 1. Same for fringe with 2 cells (2^2 configurations with a
probability of 1/(2^2) each) and other with 10 cells (2^10
configurations with a probability of 1/(2^10) each). But how is this
related to the prior distribution of <0.2,0.8> of each cell and the
product in Eq. 13.15 for the prior of one configuration?

Using combinatorics I have n!/( k!(n-k)! ) combinations to distribute k
holes over n cells, e.g. for distributing 3 pits over 15 cells
15!/(3!x12!) = 455 combinations.
Considering the 5 possible models in Figure 13.7, we have three cases
for other:
3 pits in the fringe: no pit in other - 10!/(0!x10!) * 0.2^0 * 0.8^10
2 pits in the fringe: one pit in other - 10!/(1!x9!) * 0.2^1 * 0.8^9
1 pit in the fringe: two pits in other - 10!/(2!x8!) * 0.2^2 * 0.8^8

However, these values do, of course, not sum up to 1. Even adding the
case, that is not shown in Figure 13.7 with three pits in other does not
correct that, neither does the summing over all 8 possibilities of pits
in the fringe. Actually the above equations do only fulfill our goal of
having the sum of other equal to 1, if we extend those three cases to
distribute every possible number of holes (with k ranging from 0 to 10
and n being 10) over the cells in other. Am I missing something out
here, e.g. a normalization, or am I on a completely misleading way?
So the fundamental question is:
"Is sum_other P(other) = 1?", "Why is it one?"
and, furthermore, is the result of the summation relevant at all or
could it just be put in the normalizing constant just as we did with
P(known)?

Despite my lousy explanations, I hope that my concerns are
understandable and I hope to receive some comprehensible answers.

Best regards
Dirk

Hey anybody know where to find the errata list for 2nd edition 2006
reprint (LPE, Pearson Education)ISBN : 81-7758-367-0

Regards
Abhishek Vaid

--- In aima-talk@yahoogroups.com, "abhi_944" <abhi_944@...> wrote:
>
> Hey anybody know where to find the errata list for 2nd edition 2006
> reprint (LPE, Pearson Education)ISBN :

Well, I have the 9th print of the 2nd edition, and when I checked the
errata on:

http://aima.cs.berkeley.edu/errata.html

Printing 3 or 4 (the latest on this page), all the errors has been
corrected in my book. This implies that either:

- all the errors has been found and corrected in following prints,
- no one updated the above errata with new submissions.

In your case I would do the same, compare errors mentioned in the
above errata with your book. If everything is corrected, then I guess
you have the latest. The errata applies to both the official and the
international version (which is what you seems to be having).

Sharkie

Will there be a next edition of the AIMA text book?

Yes, there will be.  The authors are discussing the process of writing
a third edition now, but don't yet have a schedule.

-Peter Norvig

On 10/1/07, per.nyblom <per.nyblom@...> wrote:
> Will there be a next edition of the AIMA text book?
>
>
>
>
> Yahoo! Groups Links
>
>
>
>

Peter Norvig wrote:
>
> Yes, there will be.  The authors are discussing
> the process of writing a third edition now,
> but don't yet have a schedule.
>
> -Peter Norvig
>
> On 10/1/07, per.nyblom <per.nyblom@...> wrote:
>> Will there be a next edition of the AIMA text book?
>>

It would be nice if future editions of the AIMA textbook
were to include some treatment of the various independent
AI projects that are out there (on the fringe?) nowadays.

http://mind.sourceforge.net/Mind.html in JavaScript
is an AI Mind tutorial program that demonstrates the
important technique of "spreading activation" at work.

http://AIMind-I.com by Mr. Frank J. Russo is a program
in Forth spawned by the SourceForge Mind project but
running independently and now able to receive e-mails.

http://mentifex.virtualentity.com is a site devoted to
Wikipedia-based open-source artificial intelligence --
the idea that Wikipedia is where students of AI may
not only learn the multidisciplinary subjects needed
for creating true AI, but may turn around and write
contributions to the very articles being studied.

This is with regards to AIMA 2nd Ed, page 36.

The description says that if a performance penalty of one point is
assessed, the agent will fare poorly. I'm at a bit of a loss to explain
how the agent could ever perform well in that situation. If, as the first
bullet indicates, the agent gets one point at each time step for a clean
square, and loses a point for each movement, wouldn't that mean the agent
could never possibly score higher than 1 or lower than 0 at the end of its
run if a peformance penalty is assessed?

(This same scenario is referenced in question 2.9, but I believe my
question only asks for clarification, not an answer.)

..Chuck..


P.S. The aima-java class TrivialVacuumEnvironment.java awards 10 points
for a clean square, rather than the one point that seems warranted based
on the rules outlined on page 36. Which is correct? I emailed Ravi Mohan
about that, but never received a response.


--
"The idea that any one of us [presidential candidates] can bring about
this change is a fantasy, it is not the truth! We need you to bring about
the change on all these issues, we need you involved, we need you taking
responsibility!"
		 --John Edwards