Jan Bogaerts - Jul 8, 2011:

Understanding how people act overcomes that impasse.

Good point.

This isn’t really true. Replace the people with computers, and an algorithm will come to the same conclusion.

Andrew
(as my own name in English) Thanks !
I just saw it, but I think there must be more out there, the type of learning I’ve seen is not what I meant, I meant the capability to make inference over observations, and find rules at their own, like a mosquito’s brain when seeking blood in the wild, there is no reasoning, but true goal seeking, by any means, I think this is the most successful approach, not a bunch of cross linked complex rules like this thing presumably is.

Robert, good point!
I open source my thoughts (which is more than my code), not my own code because I want to live from my work outcome, the model is to license code not to sell ideas, (no one pays fro them) and no foundation, university nor science hall supports my daily living costs!
Also I am sick of open source with poor or none documentation, and building the docs or my own projects for others than myself is too complex and time consuming and has no reward at all, unless facilitating others future work, I’m not so philanthropic yet! (I’m a little selfish now) But some time ago, I gave back many open source code too many times, also anonymously, and even did never get thanked, even worse I’ve got thrown out of some commercial project as soon as my (good) ideas were coded by me and successfully tested, so they did not need me any more to do money with them, so those were my last contributions to the open source cause! (at least for now) and those are the market rules, i didn’t make them!

AndyHo - Jul 10, 2011:

I just saw it, but I think there must be more out there, the type of learning I’ve seen is not what I meant, I meant the capability to make inference over observations, and find rules at their own, like a mosquito’s brain when seeking blood in the wild, there is no reasoning, but true goal seeking, by any means, I think this is the most successful approach, not a bunch of cross linked complex rules like this thing presumably is.

@AndyHo If PLOW isn’t making inferences from observations in your view, then I think that you and I must understand something completely different from those words.

If it’s a mechanical mosquito (or something similar) that you wish to create then I’d recommend that you do some reading about “subsumption architecture” (if you have not done so already) which is the basis of most robotics software these days.

The idea behind this approach is that intelligent behaviour can emerge from purely reactive interacting mechanisms, without the need to maintain an internal logical model of the world. Many script based chatbots actually work this way, though I think it is more by accident than by design.

http://en.wikipedia.org/wiki/Subsumption_architecture

AndyHo - Jul 10, 2011:

Andrew (as my own name in English) Thanks!

@AndyHo Do you mean by this that you would prefer to be addressed as “Andrew” which is also my preference? Will certainly do so from now on if that is the case.

@Andrew Andy is just fine, I have been bombed with all sort of nicknames up to AndySoft, which I inherited unintentionally ¿guess why?) all is related with my complicated-to-pronounce German family name Hohendahl, and Andy was used by my family and friends, here (academically speaking) they call me Andrés which is my real name, but Andy is Just-Fine for friends and colleges as you all.

I did knew about Act-R (which seems to be listed in the cognitive list of this Wiki entry) but this chaos-layered shells (I must admit) is not for me, at least at this time, I researched chaotic and ergodic systems, non integer space dimensions (fractals) and many other strange things, but for fun, the potential of an emerging cognitive property out of this is for me, very uncertain and not worth to research. My time also is limited, we are all born with a finite number of ticks, so I try to use them wisely… but thanks for the lead !

I do refer to the behavior as a dialog-act automaton, to allow cognitive exchange of AI-based-machines with humans in a productive way, I am certainly not addressing a ‘smart-ass’ as I stated above, but the dialog dynamics involved in turn-taking, and the related grounding decisions, anaphoric o-references constructions are rather interesting to me for modeling a successful chatbot, not a mind, but resembling the way a mind interacts with specific purposes.

@Robert,

Why would an algorithm worry about being in prison for the rest of its existence.  It seems your view of them already puts them in a prison.

How long would a computer wait on the other computers before it could conclude that there was more to the problem than just logic or at least that it now had the additional information to determine they all were marked red?  A couple of microseconds?  Computers can figure out all the possibilities much faster than you can.  I maintain that the most direct calculations can determine is that there are at least two of the three marked red. The taps don’t tell enough to declare all three are marked red.

I’ve never heard of an algorithm that understands what other machines are doing without sensors or instrumentation as agents inside the other machines.  Are you suggesting all computers have some kind of built-in protocol so they can handshake these things without arranging any direct communication paths?  If so, why wouldn’t they all three quickly know the answer and be set free?

Post deleted

I simply needed more editing time,
than was left on the EDIT timer.
So, I deleted my post for now.

See if you can figure out what these seven words all have in common?

1. Banana
2. Dresser
3. Grammar
4. Potato
5. Revive
6. Uneven
7. Assess

No, it is not that they all have at least 2 double letters….

How would Watson figure this one out?

Got it.

Could there be an algorithm for such a puzzle.  Seems to me this is reverting back into having a whole bunch of patterns to match to work on the various kinds of problems.  Just that the patterns are ways to solve puzzles instead of text to match in inputs.

Thats how I solved it. I ran through a whole bunch of tests in my mind until I found one that worked.

Given that it is a common enough kind of problem that it has its own name in the English language (see the PM that I sent you), it probably could be solved that way by a not very bright program.

See if you can find an example of a kind of problem that isn’t so common and we’ll see what we can make of it.

@Gary:

This is why I look up to all of you folks. All I see in the list of words that they ALL have in common is that they’re all in the dictionary. But then again, my grasp of the finer points of the English language (not to mention some types of reasoning skills) leave something to be desired. so in the interest of educating those not as well gifted in certain areas, would you care to enlighten us about any other commonalities those words share?

Also, in the interests of keeping the discussion “on topic”, rather than trying to admonish the participants, or try to steer the discussion back to the original topic, I’ve taken the expedient route of re-naming the thread. It ain’t much, but at least I’m TRYING to contribute!

I sent the answer to Andrew to show him that he didn’t solve it although he thought he did. And it is not so common to be easily solved by a not very bright program. Although a program could make the same mistake if it used simple genetics on typical solving techniques without going too deep into the generations.

Sorry Andrew for spoiling it for you.

They are all in the dictionary, but unlike any other word in the dictionary they all have one unique property in common. (Unless you include proper names like Ghazzah and Brenner, or non-english words like caressera, gareggera, matarrata and paressera for example.)

Good move changing the topic title.

EDIT: Just saw your last post Gary. I haven’t received your PM yet, but I look forward to learning how I was wrong ... and learning something new.

They all score at least 12 points in Scrabble?

Just kidding. I did check it out, though, and they range in letter score (no bonuses) from 6 points (Assess) to 12 points (Grammar and Revive), with the rest scoring 8 points each. Funny, some things folks will do when they’re in one of those moods!