| |
| |
| |
|
Comments:
<0> good intuition Magnus_RM. a mechanism has no need for sensation. consciousness etc. but not easy to realize for many
<1> yeah
<2> anybody awake
<1> probably more than half the population of the Earth
<1> And I am one of them :)
<2> congratulations, you're unique, just like half the people on the earth
<1> I guess so, unless I am having an exceptionally vivid dream right now.
<3> it's not unique if half of the people on earth are the same..
<2> that was kind of the joke, icez..
<1> See, another guy awake :)
<3> :]
<2> a modification of a very old joke
<2> anyway, i've been wondering about the feasability of making a program that would make predictions about play calls in football
<1> what is a "play call"?
<2> err, sorry, american football
<3> the one with the weird ball :D
<1> ok, but still, what is a play call?
<2> a play call would be like a run over the left end, or a p*** up the middle
<1> ok
<1> (foot-oval)
<1> :)
<2> anyway, there wouldn't be all that many factors involved
<1> i think it is definitely feasable
<2> i'm trying to decide if the program would just need data from previous plays in that game, or for previous games, too
<3> i think it should learn what to expect from other games..
<2> but then it wouldn't be useful until at least the second game of the season
<2> or even interesting, for that matter
<3> well..that's how it goes
<3> :P
<2> besides which, the call a coach might make against one team might differ from the call he might make in the same situation against another team
<1> i had the idea of a new kind of football (soccer) which could be called computer football, where the players get quick instructions on the fly from a computer (one for each team) which coordinates the players of a team, makes estimations of what can be done and what can not be done, analyses the situations and makes strategies which it then executes with the help of the "remote controlled" players.
<1> The players should be trained to be very alert to the on the fly instructions of the computer, but also take own initiatives withn reasonable limits. This kind of football could be very facinating to watch.
<2> i've heard of robot soccer being done before
<3> :\
<1> One problem is how the instructions should be conveyed to the players, because letting them have earphones, even specially made ones, coudl be quite dangerous, because of the roughness on the field.
<1> but i am talking about real humans, not robots
<2> oh ok
<1> so it would look just like real football when you watch it on TV but with the difference that the players would do really good and smart things
<2> i misunderstood what you were suggesting
<1> ok :)
<2> well, i doubt most people understand the good plays that are made already
<1> The computer could even instruct the players if it is worth to pretend that he was tackled in an unjust way or not, depending on what the computer thinks the judges saw or did not see. The computers (one for each team) would be connected to cameras etc, supervising the whole arena. It would also make very intelligent strategies about how to use the limited energy that each player has.
<1> So sometimes it would say "stop", to tell a player "dont run after that ball, you will just waste your energy, because you will not be able to catch it before it p***es the side line."
<2> that sounds...insanely complicated
<1> yes, quite :) but fun, if one can get it working
<1> i would love to see such a match
<1> ok, need to go to bed here
<1> bye for tonight
<4> chessguy: the program would certainly need data from previous games, otherwise where will it get its training?
<4> not near enough training data in one game
<5> lets get the chat going here
<5> dont force me back to newsgroups
<3> paros: nobody's stopping you
<5> ok
<5> I am an acolyte of my two masters
<5> One of my masters is named Rolf Pfeifer
<5> the other Gerald Edelman
<4> not this again
<5> many years ago edelman and pfeifer met in california at a place called NSI
<4> we've heard this before
<4> something new please
<5> this is all new
<5> in the bible, written by pfeifer, he mentions edelman at length
<3> yeah
<3> you already said that
<3> :\
<5> no i have never talked about pfeifer's book
<4> the religion metaphor is more fitting than you realize perhaps... the dogma and your intolerance of other views
<2> what book are you referring to as the 'bible'?
<5> Understanding Intelligence by pfeifer and scheier
<5> intolerance of other views
<5> what views would that be?
<5> athena you act as if you FILL this channel with lots of "views"
<5> which you have not
<5> I havent really ever seen you present a single 'view' here
<4> it isn't about me... i'm talking your habit of coming here to either (a) lecture about two certain AI researchers or (b) indulge your irrational animosity towards the singularity institute
<4> you're obviously intelligent, which makes this all the more sad
<5> athena i think it would be better if you responded to my posts here as they come
<5> rather than tainting them with my past behaviors
<5> hi kinoc
<5> Captain_Fourier
<6> hi paros
<5> hi
<5> hey Chromix
<5> hi vIkSiT
<7> lo paros
<5> vIkSiT what is your interest in AI?
<7> paros, a more pertinent question, is whats yours?
<5> mine is strong AI
<7> hmm, thats rather broad
<5> well actually i can be more specific
<5> memory and behavior
<5> hi oogaw
<8> in xyz + st.. what that does the s and t stand for?
<8> ID_INLINE idVec5::idVec5( const float x, const float y, const float z, const float s, const float t ) {
<8> oh speed and time
<8> how long it will take you and what speed to get there?
<5> what the f__ is this dmiles?
<8> its an AI game i am working of for Quake4
<8> so you can player doom/quake from/inside Cyc
<5> dmiles The whole AI thing in those games is all about copying human paths
<8> and cyc can control the bots
<5> otherwise the player jerks around in random movements that do not look natural
<8> yeah.. i want to make bots copy all sorts of human actions
<8> and arrange the order they play out the parts
<8> parts/action
<0> .
Return to
#ai
or
Go to some related
logs:
#perl
OSError: libLoadLibrary.so
gxine ubuntu sound problem
sovrad
#gentoo
highest value array maximum inurl:python
#math
#javascript
fedora iaxmodem rpm
Adaptec Embedded SATA HostRAID 6300ESB Linux
|
|