aima-talk : Messages : 830-860 of 946

Hi Nami, again. I think you are mixing the point I am emphasizing and Mr. Norvig does. I definitely agree with the Mr. Norvig's performance measurement criteria:

"As a general rule, it is better to design a performance measure according to what one
actually wants the in the environment, rather than according to how one thinks the agent
should behave"

I think I didn't say something against to this criteria. however, let's delve into this.

I want to change the enviroment, yes this is my goal.

Q - and what do I have to change the environment?
A - I have my agent

Q - What does my agent have to change the environment?
A - It has actions to change the environment.

Q - What actions does my agent have to change the environment ?
A - It has A,B,C and D.

Q - Can my agent have another actions to change the environment other than the ones I specified above? for example, an E action.
A - No, it can't.

Q - So, should I specify my goal according to my actions so that the environment will change the way that maximizes my expactation/goal?
A - Yes, Exactly.

Q - But wait, there might appear some side effects stemming from those 4 actions.
A - In that case you should have specified it among your actions.

As It is seen, I didn't claim anything against that performance measure criteria. so giving it as example to refute my claim is kind of off the topic.

[[ In the earlier example I gave, the flaw was a good example, as rather than looking at "How
much has the agent travelled?", they were looking at "How much has the ball rolled"... and ]]

you should consider your agent's afore specified actions , not its human point of view implications; in either case [say it travelled or rolled the ball] your agent was *MOVING* [definition: to excitite engine with some current] to some direction to maximize your goal [ *MOVE* forward as much as possible ]. In this example, as i said before, your agent only carried out the actions you stated before, it *MOVED* and found a lump to maximize its goal. however, if it had carried out another action that would be analogous to my example; such as flying, or bouncing [which you hadn't specified before the experiment].

This example of Mr norvig is similar to teacher someone to play football with this action:

Action:
You can take the ball anywhere you want excecpt the border lines.
You will use ONLY your feet to take the ball.

Goal:
Throw the ball opposite team's goalpost as much as possible, do not let them do the same thing and do not hurt them.

Result:
Players use ONLY their feet -as expected - and you punish them since they don't use their HEADs to score-which you didn't include in the action list. This is so ridicilous, teacher should have told the rules better; this is his own fault not the players.

--- 23/06/08 Pzt tarihinde daashmashty <nima.talebi@...> şöyle yazıyor:

Kimden: daashmashty <nima.talebi@...>
Konu: [aima-talk] Re: A serious logical error in chapter 2.
Kime: aima-talk@yahoogroups.com
Tarihi: 23 Haziran 2008 Pazartesi, 11:52

You're argument is that dumping is not in the action list - but that's not the point that's
being made here.

The key that is being conveyed here is this...

"As a general rule, it is better to design a performance measure according to what one
actually wants the in the environment, rather than according to how one thinks the agent
should behave"

----

In the earlier example I gave, the flaw was a good example, as rather than looking at "How
much has the agent travelled?", they were looking at "How much has the ball rolled"... and
the robot discovered an honest solution to make the ball roll faster, which was not what
the designers were looking for.

Analogously here, we're saying that the agent could find a similar `flaw' (from the
designer's perspective) in the reward system, and continually dump and suck - this is just
an observation. .. We are hypothesizing. .. as in "Imagine if..."
...imagine *if* the robot could *dump*, then you'd see that this suck-dump-loop could be
a possibility, and to not allow that to happen, so we need to be careful about how the
reward system is set up.

I hope that clarifies it.

Nima

--- In aima-talk@yahoogrou ps.com, ozgur aydogan <aydogan_ozgur@ ...> wrote:
>
> Hi thanks for the answer, but , frankly, I know what Mr. Norvig wants to point by that
exceptional action of the cleaner. And, I see your example as well. But there is a little
difference between yours and the one I am mentioning about; your robot carried out the
actions you stated at the beginning of your experiment. It went through a direction[must
be only action you defined for it?] looking for the maximum goal you had described. Thus,
it only carried out the action you wanted it to do[and that scratching dog like position was
a result of your goal and action definitions. ]
>
> if you excuse me, I want to exemplify my question:
>
> Firstly, let's forget about the agent and just focus on the actions of it to abstract the
example. Here are the actions:
>
> A, B, C, D and those are mutually exclusive and none of them has side effects[like
DUMPing].Â how would you get an extra action [say it E] out of this four ones? you can't
unless you implied it[Unlike your robotic example, It only carried out the actions you
described initially]. Although this looks trivial issue, it lacks information and bad example
for an introductory example. Lastly, If I added an additional action of a Truth Table during
the exam, my teacher would definitely have me failed :S
>
> Thanks again.
>
>
>
>
>
> --- 22/06/08 Pzr tarihinde Nima Talebi <nima.talebi@ ...> ÅŸÃ¶yle yazÄ±yor:
> Kimden: Nima Talebi <nima.talebi@ ...>
> Konu: Re: [aima-talk] A serious logical error in chapter 2.
> Kime: aima-talk@yahoogrou ps.com
> Tarihi: 22 Haziran 2008 Pazar, 5:42
>
>
>
>
>
>
>
>
>
>
>
> I think the mention of `dump', not being in the action list of the agent, as you
correctly stated, in this environment at least serves no purpose.
> I think (please correct me if I'm wrong), Dr. Norvig mentions the action of a `dump' only
as a means of driving a point home - If allowed to be the designer of it's own performance
measure, it could easily make life easy for itself and, as an *example*, enter a loop of
clean-->dump- ->clean-- >dump... and satisfy itself that it is performing extremely well,
while in reality, it's doing nothing.
>
> Here is another example I've comeÂ acrossÂ which may help...
> There was a robotics experiment (reinforcement learning) where a robot had a ball-
mouse attached to it behind, and it was `rewarded' for how much the ball rolled
*forward*, then allowed to experiment in a room... hopefully to learn to move as fast as
possible by itself.
>
> It was left to explore overnight, and in the morning, it was found sitting at an area in
the room where there was a lump on the ground... and what the agent was found doing
was this...
>
> Best way to draw a picture is this - imagine a dog rubbing it's behind on the carpet as to
scratch it.
> In effect - that is what it had learned to do. It was (incorrectly) being rewarded because
the series of motions it had learned and was carrying was indeedÂ honoringÂ (cheating)
the reward system...
> Â * the ball was rolling forward faster than ever before.
> This isÂ analogousÂ to the vacuum cleaner taking a dump and sucking it backup. Â I
hope I've answered your question? =)
>
> Nima
>
> On Sun, Jun 22, 2008 at 10:58 AM, aydogan_ozgur <aydogan_ozgur@ yahoo.com>
wrote:
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
> In Performance Measures of Chapter 2, Mr. Norvig refutes the goal of
>
> counting the number of cleaning a square by claiming that the
>
> cleaner,in which case, can successionly clean and DUMP the dirt of
>
> the same square, which is clearly worthless.
>
>
>
> However, there is a nontrivial glitch in this proposition. The Actions
>
> of the cleaner are
>
>
>
> to move right,
>
> to move left,
>
> to clean or
>
> to do nothing.
>
>
>
> In other words, there is no DUMP action, which Mr. Norvig uses to
>
> refute the proposition. So, I think, he can't just make up another
>
> action to refute the proposition. I hope, I am not exagerrating the case?
>
>
>
> Thanks for reading.
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
> ____________ _________ _________ _________ _________ _________ _
> Yahoo! kullaniyor musunuz? http://tr.mail. yahoo.com
> Istenmeyen postadan biktiniz mi? Istenmeyen postadan en iyi korunma
> Yahoo! Posta'da
>

Yahoo! kullaniyor musunuz?
Istenmeyen postadan biktiniz mi? Istenmeyen postadan en iyi korunma Yahoo! Posta'da
http://tr.mail.yahoo.com

Reply | Messages in this Topic (5)

#831

From: "takeforcemanager" <takeforcemanager@...>
Date: Wed Aug 6, 2008 8:19 am
Subject: Timetable as CSP

takeforceman...

Hi all!
There is an exercise, 5.5(b) which I'm stuck with. I just can't get a
correct approach for it...
Consider we have 3 Groups (GroupA, GroupB, GroupC), 3 Rooms (RoomA,
RoomB, RoomC), 3 Teachers(TeacherA, TeacherB, TeacherC), 3 subjects
(SubjA, SubjB, SubjC) and 3 possible timeframes(Lesson1, Lesson2,
Lesson3).
Then I put constrains, that link teachers to subjects:
teach(TeacherA, SubjA)
teach(TeacherA, SubjC)
teach(TeacherB, SubjB)
teach(TeacherC, SubjC)
These are binary relations, that's excellent.
And now, how to write a constraint, that says: "Only one group can be
in the classroom" or "Only one teacher can teach one subject on the
same lesson" or "Group can have only one subject in the same lesson".

I just can't get an idea how to express that. What should be
variables? For example, we can put variable TimetableRecord, that is a
record and has the following fields: Group, Classroom, Teacher,
Lesson, Subject. But that is nit correct, I know... What approach is
correct?

I know, answers to exercises can';t be post here, but I don't need it.
I just need a hint on how to express restrictions I mentioned above...

#832

From: "Brandon Corfman" <bcorfman@...>
Date: Thu Aug 21, 2008 6:08 pm
Subject: AIMA code for planning?

bcorfman

I notice there is no AIMA Python or Lisp code for any of the planning
algorithms in chapter 11. I'd really like to find out if anyone has
developed any planning code based on the book pseudocode before I
strike out on my own.

Please post a quick reply if you've developed any  planning code that
I could look at or use. Thanks.

Best regards,
Brandon

#833

From: "helpathand1234" <helpathand1234@...>
Date: Fri Sep 5, 2008 11:36 am
Subject: help on evaluation in dynamic bayesian network - DECAYED MCMC.

helpathand1234

Sir,
      In your book AIMA, in the chapter of dynamic bayesian
networks(sample umbrella network),say; if I wish to
calculate the probability of rain on day 5, given evidence of rain
from day 1 to 5 i.e. p(X5|E1:5), using DECAYED-MCMC,
how do I implement the formula( inverse polynomial decay-
{g(t)=(alpha)(T-t+1)e-(1+delta)} given in
Mr. Stuart Russell's UAI-02 paper on decayed MCMC) for the above
problem ?

	 Please do help me,
Waiting for your reply,
Regards,
Abhijit.

#834

From: "Moeljono Widjaja" <moeljono.widjaja@...>
Date: Tue Sep 9, 2008 5:51 am
Subject: Error in uploading aima.lisp

mwidjaja2000

Dear All:

Would you please help me solving this problem?
I am trying to upload the lisp codes (aima.lisp) but I encountered a problem.
I pasted the error message in this email.
I have tried to look for information to solve this problem via search engine or archive aima mailing list, but no luck so far.

FYI, I am using Ubuntu 7.10 and install the aima-lisp package from Ubuntu repository.

Thank you for your help.

--
Regards,

Moeljono Widjaja
Jakarta - Indonesia

===============

moeljono@narnia:~/aima$ clisp
i i i i i i i       ooooo    o        ooooooo   ooooo   ooooo
I I I I I I I      8     8   8           8     8     o 8    8
I \ `+' / I      8         8           8     8        8    8
   \ `-+-' /       8         8           8      ooooo   8oooo
    `-__|__-'        8         8           8           8 8
        |            8     o   8           8     o     8 8
------+------       ooooo    8oooooo ooo8ooo   ooooo   8

Copyright (c) Bruno Haible, Michael Stoll 1992, 1993
Copyright (c) Bruno Haible, Marcus Daniels 1994-1997
Copyright (c) Bruno Haible, Pierpaolo Bernardi, Sam Steingold 1998
Copyright (c) Bruno Haible, Sam Steingold 1999-2000
Copyright (c) Sam Steingold, Bruno Haible 2001-2006

[1]> (load "aima.lisp")
;; Loading file aima.lisp ...
*** - READ from #<INPUT BUFFERED FILE-STREAM CHARACTER #P"aima.lisp" @150>:
       #<PACKAGE COMMON-LISP-CONTROLLER>
      has no external symbol with name "SOURCE-ROOT-PATH-TO-FASL-PATH"
The following restarts are available:
ABORT          :R1      ABORT
Break 1 [2]>

#835

From: "inyourmind6684" <dtamayo@...>
Date: Tue Sep 16, 2008 5:51 am
Subject: Problems using aima-java

inyourmind6684

I'm having trouble opening up the aima-java. I am using eclipse to
open the .project file. However, everytime I try to do this, I just
open up a blank workspace. There is no project defined and I have no
idea why. The very first time I did this, I saw the project and
everything. Then after that I seem to have lost the ability to view
it. I have tried unpacking the zip several times and even reinstalling
the eclipse program several times to make sure the settings were still
the same and have yet to be able to view the project again. Can anyone
help with this problem? Thank you for your time.

#836

From: "Ravi Mohan" <magesmail@...>
Date: Tue Sep 16, 2008 8:01 am
Subject: Re: Problems using aima-java

magesmail

If you are having trouble with .project file, delete it. Use the
"Import project from source" option in eclipse and it will generate a
new .project file for you.

Cheers,
Ravi

--- In aima-talk@yahoogroups.com, "inyourmind6684" <dtamayo@...> wrote:
>
> I'm having trouble opening up the aima-java. I am using eclipse to
> open the .project file. However, everytime I try to do this, I just
> open up a blank workspace. There is no project defined and I have no
> idea why. The very first time I did this, I saw the project and
> everything. Then after that I seem to have lost the ability to view
> it. I have tried unpacking the zip several times and even reinstalling
> the eclipse program several times to make sure the settings were still
> the same and have yet to be able to view the project again. Can anyone
> help with this problem? Thank you for your time.
>

#837

From: "inyourmind6684" <dtamayo@...>
Date: Tue Sep 16, 2008 9:28 pm
Subject: Re: Problems using aima-java

inyourmind6684

At what level should I import the source code? should I just import
the aima level? or the src sublevel? Thank you for your time.


--- In aima-talk@yahoogroups.com, "Ravi Mohan" <magesmail@...> wrote:
>
>
> If you are having trouble with .project file, delete it. Use the
> "Import project from source" option in eclipse and it will generate a
> new .project file for you.
>
> Cheers,
> Ravi
>
> --- In aima-talk@yahoogroups.com, "inyourmind6684" <dtamayo@> wrote:
> >
> > I'm having trouble opening up the aima-java. I am using eclipse to
> > open the .project file. However, everytime I try to do this, I just
> > open up a blank workspace. There is no project defined and I have no
> > idea why. The very first time I did this, I saw the project and
> > everything. Then after that I seem to have lost the ability to view
> > it. I have tried unpacking the zip several times and even reinstalling
> > the eclipse program several times to make sure the settings were still
> > the same and have yet to be able to view the project again. Can anyone
> > help with this problem? Thank you for your time.
> >
>

#838

From: "kiana" <blacksilk79@...>
Date: Wed Sep 24, 2008 8:53 pm
Subject: Chapter 3 discussion

blacksilk79

I understand the material in the chapter. However, the exercises are
completely confusing to me. Therefore, a discussion of the chapter may
clarify. For the 8-puzzle problem, I am not able to understand the
existence of two disjoint sets of all possible states in which a state
from one set can't transform to a state in the other set by any number
of moves. I would assume any state can be reached by in other state. In
other words, there are situations (initial state - goal configuration)
in which the puzzle isn't solvable? Does the two disjoint sets exit
because the problem becomes an NP-Complete problem for those states
trying to reach the states in the other disjoint set. And, the 9!/2
calculation for all possible states derives from this theorem of
exactly half of the possible states transform into a given goal, is
this true?

#839

From: "kiana" <blacksilk79@...>
Date: Wed Sep 24, 2008 8:38 pm
Subject: Re: Fig. 3.4 Edition 2

blacksilk79

--- In aima-talk@yahoogroups.com, ramanathan pl <ramkrshn36@...>
wrote:
>
> Hi,
>
> When I was going through the 8 puzzle example in fig. 3.4 of AIMA
2nd edition, I found that from the given start state (7 2 4 5 0 6 8 3
1), it is NOT possible to reach the given Goal state ( 0 1 2 3 4 5 6
7 8) since both the states belong to two disjoint sets and it is not
possible to transform from a state in one set to a state in another
set , immaterial of how many moves we make.
>
> Any similar comments and discussions appreciated.
>
> Thanks
> Ram
>
>
>
>
>
It's evident two disjoint sets of all possible states for the 8-
puzzle problem exist. However, I don't understand. I would assume it
is possible to reach any state from a given state. What are the
differences between the two disjoint sets? I am seriously confused.
______________________________________________________________________
_________
> Be a better friend, newshound, and
> know-it-all with Yahoo! Mobile.  Try it now.
http://mobile.yahoo.com/;_ylt=Ahu06i62sR8HDtDypao8Wcj9tAcJ
>

Reply | Messages in this Topic (2)

#840

From: "kiana" <blacksilk79@...>
Date: Wed Sep 24, 2008 9:25 pm
Subject: Re: Chapter 3 discussion

blacksilk79

--- In aima-talk@yahoogroups.com, "kiana" <blacksilk79@...> wrote:
>
> I understand the material in the chapter. However, the exercises
are
> completely confusing to me. Therefore, a discussion of the chapter
may
> clarify. For the 8-puzzle problem, I am not able to understand the
> existence of two disjoint sets of all possible states in which a
state
> from one set can't transform to a state in the other set by any
number
> of moves. I would assume any state can be reached by in other
state. In
> other words, there are situations (initial state - goal
configuration)
> in which the puzzle isn't solvable? Does the two disjoint sets exit
> because the problem becomes an NP-Complete problem for those states
> trying to reach the states in the other disjoint set. And, the 9!/2
> calculation for all possible states derives from this theorem of
> exactly half of the possible states transform into a given goal, is
> this true?
>
I'm confused, any clarification is greatly appreciated.

#841

From: "kiana" <blacksilk79@...>
Date: Wed Sep 24, 2008 9:20 pm
Subject: Re: Chapter 3 discussion

blacksilk79

--- In aima-talk@yahoogroups.com, "kiana" <blacksilk79@...> wrote:
>
> I understand the material in the chapter. However, the exercises
are
> completely confusing to me. Therefore, a discussion of the chapter
may
> clarify. For the 8-puzzle problem, I am not able to understand the
> existence of two disjoint sets of all possible states in which a
state
> from one set can't transform to a state in the other set by any
number
> of moves. I would assume any state can be reached by in other
state. In
> other words, there are situations (initial state - goal
configuration)
> in which the puzzle isn't solvable? Does the two disjoint sets exit
> because the problem becomes an NP-Complete problem for those states
> trying to reach the states in the other disjoint set. And, the 9!/2
> calculation for all possible states derives from this theorem of
> exactly half of the possible states transform into a given goal, is
> this true?
>
Discussions of the topics in the book leads to a better understanding
of the material. I need clarification of the topics in order to
answer the exercies, not necessarily the answer to the questions. In
order for me to understand the material further, I need to understand
the questions. The existence of the disjoint sets were discussed in
book, so I would like to elaborate on that topic. I don't understand
that section of the book. Thank you.

#842

From: "inyourmind6684" <dtamayo@...>
Date: Thu Sep 25, 2008 6:18 am
Subject: Problems with AIMA code QueueSearch

inyourmind6684

I'm looking to create a test file and successor function for BFS in
the AIMA code. However, in order to instantiate a BFS I have to pass
in a QueueSearch. I ran into the problem when trying to instantiate
the QueueSearch. I know that QueueSearch extends the NodeExpander but
I've tried several ways to instantiate it and I keep getting
complaints from eclipse. Does anyone know how to create a QueueSearch
so that eclipse doesn't complain?

Thank you for your time.

#843

From: "Ravi Mohan" <magesmail@...>
Date: Thu Sep 25, 2008 9:32 am
Subject: Re: Problems with AIMA code QueueSearch

magesmail

The code for QueueSearch begins as follows

"public abstract class QueueSearch  .... "

The "abstract" keyword implies you cannot instantiate the class. You
need to instantiate one of its concrete subclasses (TreeSearch or
GraphSearch).

The demo and unit test classes have a dozen plus examples of how these
may be instantiated. The readme file explains the two hierarchies of
search and how they relate to each other.

"> I've tried several ways to instantiate it and I keep getting
> complaints from eclipse."

When this happens the next time, do try to understand what the error
message is trying to communicate. An attempt at doing this would have
exposed the "trying to instantiate an abstract class" error.

In other words, eclipse complains for a reason!

--- In aima-talk@yahoogroups.com, "inyourmind6684" <dtamayo@...> wrote:
>
> I'm looking to create a test file and successor function for BFS in
> the AIMA code. However, in order to instantiate a BFS I have to pass
> in a QueueSearch. I ran into the problem when trying to instantiate
> the QueueSearch. I know that QueueSearch extends the NodeExpander but
> I've tried several ways to instantiate it and I keep getting
> complaints from eclipse. Does anyone know how to create a QueueSearch
> so that eclipse doesn't complain?
>
> Thank you for your time.
>

#844

From: "kiana" <blacksilk79@...>
Date: Sun Sep 28, 2008 7:09 pm
Subject: Re: Problems with AIMA code QueueSearch

blacksilk79

--- In aima-talk@yahoogroups.com, "inyourmind6684" <dtamayo@...> wrote:
>
> I'm looking to create a test file and successor function for BFS in
> the AIMA code. However, in order to instantiate a BFS I have to pass
> in a QueueSearch. I ran into the problem when trying to instantiate
> the QueueSearch. I know that QueueSearch extends the NodeExpander but
> I've tried several ways to instantiate it and I keep getting
> complaints from eclipse. Does anyone know how to create a QueueSearch
> so that eclipse doesn't complain?
>
> Thank you for your time.
>
Do we use the AIMA code examples, i.e. BFS and QueueSearch, for the
exercises in the book? Or, are we supposed to define and implement all
of the code for the exercises ourselves?

#845

From: "Nima Talebi" <nima.talebi@...>
Date: Thu Oct 23, 2008 2:08 pm
Subject: Re: Excercise 18.14, page 677

daashmashty

Okay, I can already see a flaw in this, here's my revised version...

I've attached my formula as a png, hope it gets through, in case it doesn't...

Text:
E=|_sum_{{i=0};{i=M-m};{|_pmatrix1x2_{{M};{m+i}}ε^{m+i}}}

LaTeX:
\mbox{E}=\sum_{i=0}^{i=M-m}{\left(\begin{array}{c} M \\ m+i \end{array}\right)\epsilon ^{m+i}}

...where m is the bigger half - that is the `half' that has just enough power to out-weight the other half, for example where M = 20, m = 11, where M = 5, m = 3.

Again, I don't think I have the right, or at least most elegant solution to the problem, I feel it has to be much simpler - otherwise the question wouldn't seriously expect us to calculate the above formula for M = 20 as that sum would expand into 9 terms!

Nima

On Fri, Oct 24, 2008 at 12:37 AM, Nima Talebi <nima.talebi@...> wrote:

Hi,

I'm interested to see if my solution is even remotely correct, the question is to derive the error formula given the number of experts/hypotheses (M) and also the error of each hypotheses (e) - assuming all to be the same, and that all are independent of one another...

I treated this as a probability problem (and unsure if I should have done so), and hence derived a formula which yields the following results...

M:5, e:0.10, 0.0100
M:5, e:0.20, 0.0800
M:5, e:0.40, 0.6400
M:10, e:0.10, 0.0002
M:10, e:0.20, 0.0134
M:10, e:0.40, 0.8602
M:20, e:0.10, 0.0000
M:20, e:0.20, 0.0034
M:20, e:0.40, 7.0448

The third column is the total error for the network I calculated; it certainly looks wrong from looking at the last value, but I'd like to know how to go about solving this as it's rather intriguing.

My `formula' in Python form is as follows...

def fact(k):
return reduce(lambda i, j : i*j, range(1, k+1))

for M in 5, 10, 20:
for e in 0.1, 0.2, 0.4:
    m = M/2+1
    _m = M-m

    print "M:%d, e:%0.2f, %0.4f"%(M, e, (fact(M) * pow(e, m))/(fact(_m) * fact(m)))

Thanks to anyone who has a crack at solving this =)

Nima

--
Nima Talebiw: http://ai.autonomy.net.au/
p: +61-4-0667-7607 m: nima@...

#846

From: "Nima Talebi" <nima.talebi@...>
Date: Thu Oct 23, 2008 1:37 pm
Subject: Excercise 18.14, page 677

daashmashty

Hi,

I'm interested to see if my solution is even remotely correct, the question is to derive the error formula given the number of experts/hypotheses (M) and also the error of each hypotheses (e) - assuming all to be the same, and that all are independent of one another...

I treated this as a probability problem (and unsure if I should have done so), and hence derived a formula which yields the following results...

M:5, e:0.10, 0.0100
M:5, e:0.20, 0.0800
M:5, e:0.40, 0.6400
M:10, e:0.10, 0.0002
M:10, e:0.20, 0.0134
M:10, e:0.40, 0.8602
M:20, e:0.10, 0.0000
M:20, e:0.20, 0.0034
M:20, e:0.40, 7.0448

The third column is the total error for the network I calculated; it certainly looks wrong from looking at the last value, but I'd like to know how to go about solving this as it's rather intriguing.

My `formula' in Python form is as follows...

def fact(k):
return reduce(lambda i, j : i*j, range(1, k+1))

for M in 5, 10, 20:
for e in 0.1, 0.2, 0.4:
    m = M/2+1
    _m = M-m

    print "M:%d, e:%0.2f, %0.4f"%(M, e, (fact(M) * pow(e, m))/(fact(_m) * fact(m)))

Thanks to anyone who has a crack at solving this =)

Nima

#847

From: "daashmashty" <nima.talebi@...>
Date: Thu Oct 23, 2008 10:51 pm
Subject: Re: Excercise 18.14, page 677

daashmashty

The attachment was stripped as suspected, so here's the readable version...

http://ai.autonomy.net.au/chrome/site/AIMA-18.4.png

Nima

--- In aima-talk@yahoogroups.com, "Nima Talebi" <nima.talebi@...> wrote:
>
> Okay, I can already see a flaw in this, here's my revised version...
>
> I've attached my formula as a png, hope it gets through, in case it
> doesn't...
>
> *Text:*
> E=|_sum_{{i=0};{i=M-m};{|_pmatrix1x2_{{M};{m+i}}ε^{m+i}}}
>
> *LaTeX:
> *\mbox{E}=\sum_{i=0}^{i=M-m}{\left(\begin{array}{c} M \\ m+i
> \end{array}\right)\epsilon ^{m+i}}
>
> ...where m is the *bigger half* - that is the `half' that has just enough
> power to out-weight the other half, for example where M = 20, m = 11, where
> M = 5, m = 3.
>
> Again, I don't think I have the right, or at least most elegant solution to
> the problem, I feel it has to be much simpler - otherwise the question
> wouldn't seriously expect us to calculate the above formula for M = 20 as
> that sum would expand into 9 terms!
>
> Nima
>
>
> On Fri, Oct 24, 2008 at 12:37 AM, Nima Talebi <nima.talebi@...> wrote:
>
> > Hi,
> >
> > I'm interested to see if my solution is even remotely correct, the question
> > is to derive the error formula given the number of experts/hypotheses (M)
> > and also the error of each hypotheses (e) - assuming all to be the same, and
> > that all are independent of one another...
> >
> > I treated this as a probability problem (and unsure if I should have done
> > so), and hence derived a formula which yields the following results...
> >
> > M:5, e:0.10, 0.0100
> > M:5, e:0.20, 0.0800
> > M:5, e:0.40, 0.6400
> > M:10, e:0.10, 0.0002
> > M:10, e:0.20, 0.0134
> > M:10, e:0.40, 0.8602
> > M:20, e:0.10, 0.0000
> > M:20, e:0.20, 0.0034
> > M:20, e:0.40, 7.0448
> >
> > The third column is the total error for the network I calculated; it
> > certainly looks wrong from looking at the last value, but I'd like to know
> > how to go about solving this as it's rather intriguing.
> >
> > My `formula' in Python form is as follows...
> >
> > def fact(k):
> >   return reduce(lambda i, j : i*j, range(1, k+1))
> >
> > for M in 5, 10, 20:
> >   for e in 0.1, 0.2, 0.4:
> >     m  = M/2+1
> >     _m = M-m
> >
> >     print "M:%d, e:%0.2f, %0.4f"%(M, e, (fact(M) * pow(e, m))/(fact(_m) *
> > fact(m)))
> >
> > Thanks to anyone who has a crack at solving this =)
> >
> > Nima
> >
> >
>
>
> --
> Nima Talebiw: http://ai.autonomy.net.au/
> p: +61-4-0667-7607  m: nima@...
>

#848

From: "sprgr77" <sprgr77@...>
Date: Sun Nov 2, 2008 7:56 am
Subject: looking for soloutions

sprgr77

Iam interested in AI .I need guidelines to solve the exercises in the
book"Artificial intelligence A modern approach".

#849

From: "daashmashty" <nima.talebi@...>
Date: Sun Nov 2, 2008 12:31 pm
Subject: Re: looking for soloutions

daashmashty

--- In aima-talk@yahoogroups.com, "sprgr77" <sprgr77@...> wrote:
>
> Iam interested in AI .I need guidelines to solve the exercises in the
> book"Artificial intelligence A modern approach".
>

...and your question would be... ?

#850

From: "Mani Sabri" <mani_sabri@...>
Date: Sat Dec 6, 2008 7:32 am
Subject: AIMA python

mani_sabri

hi
I was looking at AIMA python code repository in google and asking
myself is python still considered viable in this community. the java
repository seems to have all the recent focus and development. why?
personal preferences of project owners or some other reasons?


Best Regards
Mani

#851

From: "Ravi Mohan" <magesmail@...>
Date: Sat Dec 6, 2008 8:11 am
Subject: Re: AIMA python

magesmail

Hi,

I wouldn't want to speak for the maintainers  of the python
repository, but as the maintainer of the java code repository I think
it is important to understand how open source works.

The amount of progress is directly dependent on how much free time and
energy the maintainers have and often this has peaks and lulls. I
wrote most of the java code for AIMA, but I have been insanely busy
over the last year. Dr Ciaran O'Reilly has been adding a *lot* of
code, however, mostly in the logic section, but also filling in some
of the gaps in search.(see the svn branch dedicated to this effort.
This will be merged into  a new release soon).

Next year, (2009) I'll have more free time and I plan to add graphics
to the existing codebase. Particularly for small projects with
contributors in the single digits, this ebb and flow is just how it works.

If you believe that Python makes a good implementation language (I
think it does - the size of the python implementation of an algorithm
is a fraction of the size of the equivalent java code), please
contribute some code. That way, everyone wins.

My 2 cents,

Ravi

--- In aima-talk@yahoogroups.com, "Mani Sabri" <mani_sabri@...> wrote:
>
> hi
> I was looking at AIMA python code repository in google and asking
> myself is python still considered viable in this community. the java
> repository seems to have all the recent focus and development. why?
> personal preferences of project owners or some other reasons?
>
>
> Best Regards
> Mani
>

#853

From: "Nima Talebi" <nima.talebi@...>
Date: Sun Dec 7, 2008 1:47 am
Subject: Re: Re: AIMA python

daashmashty

Hi Mani

I've started writing a framework, however it's not necessarily because there doesn't exist a framework which I can use (maybe one exists that is good, and maybe there isn't)... My main reason is to get a deeper understanding; As Confucious said...

I hear and I forget. I see and I remember. I do and I understand.

Anyway, I'm using Objective-C and C, and the Cocoa framework. I'd like to learn how to integrate python and ObjC (I think it is possible, but have failed to get it working)... I think that would be awesome for me - as I can write proof-of-concepts in python, then convert parts to C/ObjC as I see fit.

So far I've implemented a Multi-layer (non-recurrent) NN with back-prop in python, SOM with various 1D, 2D and 3D lattices and an OpenGL view of the machine learning, and similarly a Perceptron.

I'm not sure how many people here would be interested in writing Mac software though =)

Nima

On Sat, Dec 6, 2008 at 8:45 PM, Mani Sabri <mani_sabri@...> wrote:

No doubt that python makes a good implementation language. and it will
be fast enough if we use Pyrex or Cython along with c/c++ extensions.
I wrote some code for multi-Agent realtime environments and later I
found out that there is a project named SPADE (Smart Python
multi-Agent Development Environment) which use XMPP/Jabber and
implements FIPA standards but the problem with SPADE is that there is
no community around them (no forum no mailing list just the emails of
the owners).
so I don't know where to start, I want to learn and to have a serious
framework at the end of my study, may be its not the right direction
(is it?) but if it is I will be glad to contribute to this project or
make a fork of it and ...

Best Regards
Mani

--- In aima-talk@yahoogroups.com, "Ravi Mohan" <magesmail@...> wrote:
>
> Hi,
>
> I wouldn't want to speak for the maintainers of the python
> repository, but as the maintainer of the java code repository I think
> it is important to understand how open source works.
>
> The amount of progress is directly dependent on how much free time and
> energy the maintainers have and often this has peaks and lulls. I
> wrote most of the java code for AIMA, but I have been insanely busy
> over the last year. Dr Ciaran O'Reilly has been adding a *lot* of
> code, however, mostly in the logic section, but also filling in some
> of the gaps in search.(see the svn branch dedicated to this effort.
> This will be merged into a new release soon).
>
> Next year, (2009) I'll have more free time and I plan to add graphics
> to the existing codebase. Particularly for small projects with
> contributors in the single digits, this ebb and flow is just how it
works.
>
> If you believe that Python makes a good implementation language (I
> think it does - the size of the python implementation of an algorithm
> is a fraction of the size of the equivalent java code), please
> contribute some code. That way, everyone wins.
>
> My 2 cents,
>
> Ravi
>
>
> --- In aima-talk@yahoogroups.com, "Mani Sabri" <mani_sabri@> wrote:
> >
> > hi
> > I was looking at AIMA python code repository in google and asking
> > myself is python still considered viable in this community. the java
> > repository seems to have all the recent focus and development. why?
> > personal preferences of project owners or some other reasons?
> >
> >
> > Best Regards
> > Mani
> >
>

--
Nima Talebiw: http://ai.autonomy.net.au/
p: +61-4-0667-7607 m: nima@...

#854

From: Bob Futrelle <bob.futrelle@...>
Date: Wed Jan 21, 2009 12:32 pm
Subject: Fig 3.1 vs rest of chapter?

bobfutrelle

The formulation in Fig 3.1 is not mentioned again, so its application
to the clear examples in Sec 3.2 is obscure at best.

How *does* it apply? Esp. since it only returns a single action.

    -- Bob Futrelle

Newton Mass &
Northeastern University
Sent from my iPhone

#855

From: Peter Norvig <peter@...>
Date: Thu Jan 22, 2009 5:58 pm
Subject: Re: Fig 3.1 vs rest of chapter?

norvig

You're right; Fig 3.1 is not well-integrated with the text -- but we will have better integration in the 3rd edition of AIMA.

As for how it works: the idea is that if the agent has a plan (in the variable seq), it executes the plan, one step each time it is called. If it does not have a plan, it forumlates a goal, then a problem, then searches for a plan that solves the problem, then begins executing it.

-Peter

On Wed, Jan 21, 2009 at 4:32 AM, Bob Futrelle <bob.futrelle@...> wrote:

The formulation in Fig 3.1 is not mentioned again, so its application
to the clear examples in Sec 3.2 is obscure at best.

How *does* it apply? Esp. since it only returns a single action.

-- Bob Futrelle

Newton Mass &
Northeastern University
Sent from my iPhone

------------------------------------

Yahoo! Groups Links

<*> To visit your group on the web, go to:
http://groups.yahoo.com/group/aima-talk/

<*> Your email settings:
Individual Email | Traditional

<*> To change settings online go to:
http://groups.yahoo.com/group/aima-talk/join
(Yahoo! ID required)

<*> To change settings via email:
mailto:aima-talk-digest@yahoogroups.com
mailto:aima-talk-fullfeatured@yahoogroups.com

<*> To unsubscribe from this group, send an email to:
aima-talk-unsubscribe@yahoogroups.com

<*> Your use of Yahoo! Groups is subject to:
http://docs.yahoo.com/info/terms/

#856

From: Victor Dalmagro <victor.dalmagro@...>
Date: Mon Mar 9, 2009 3:13 pm
Subject: Re: Chapter 7 - Algorithm 7.12

victor.dalmagro

Hi all!

I'm new in AI and I cannot understand this algorithm. Can anybody give me a help?

Thanks in advance!

Victor

Yahoo! Cocina
Recetas pr�cticas y comida saludable
Visit� http://ar.mujer.yahoo.com/cocina/

#857

From: "victor.dalmagro" <victor.dalmagro@...>
Date: Thu Mar 12, 2009 1:32 pm
Subject: Chapter 7 - Algorithm 7.12

victor.dalmagro

Hi all!

I'm new in AI and I cannot understand this algorithm. Can anybody give me a
help?

Thanks in advance!

Victor

#858

From: "Peter Norvig" <peter@...>
Date: Wed Apr 22, 2009 4:51 am
Subject: AIMA 3rd edition cover design contest

norvig

We're having a contest to design the cover for the 3rd edition.

Open to instructors and students, see the contest rules at
http://media.pearsoncmg.com/ph/esm/ecs_ai/ .  Deadline June 15th.

-Peter Norvig

Reply | Messages in this Topic (6)

#859

From: Donnchadh Ó Donnabháin <donnchadh@...>
Date: Wed Apr 22, 2009 9:11 am
Subject: Re: AIMA 3rd edition cover design contest

dodonnabhain

That's great. Is there any information available on what to expect in
the 3rd edition or an estimate of when to expect it to be available
for purchase?

   Thanks

     Donnchadh

2009/4/22 Peter Norvig <peter@...>:
>
>
> We're having a contest to design the cover for the 3rd edition.
>
> Open to instructors and students, see the contest rules at
> http://media.pearsoncmg.com/ph/esm/ecs_ai/ . Deadline June 15th.
>
> -Peter Norvig
>

Reply | Messages in this Topic (6)

#860

From: <futrelle@...>
Date: Wed Apr 22, 2009 12:57 pm
Subject: Re: AIMA 3rd edition cover design contest

bobfutrelle