Subject:
|
Re: Selflearning Scout robot
|
Newsgroups:
|
lugnet.robotics
|
Date:
|
Wed, 24 Nov 1999 15:02:43 GMT
|
Viewed:
|
639 times
|
| |
|
|
On Wed, 24 Nov 1999, Mario Ferrari wrote:
> Did you succeed in implementing a neural net on the RCX? I'm sure it IS
> possible, if not with NQC surely with legOS (though with NQC too we
> succeeded in emulating arrays for our tic-tac-toe player). But I haven't
> seen any code around yet.
We have something vaguely similar (a "reinforcement learning" robot,
specifically Q learning.) Probably even more computationally intense than
the average neural net (depending on a number of factors). I have some
code, but I've been meaning to wait until I can write up a proper
explanation of what the code does and hopefully convert it to fixed-point
so that it won't be so ridiculously slow. Let me know if you are interested-
Luis
#######################################################################
Profanity is the one language that all programmers understand.
-Anonymous
#######################################################################
|
|
Message is in Reply To:
 | | Re: Selflearning Scout robot
|
| (...) Thanks! Excuse me, I couldn't resist poking you about that :-) (...) think (...) Did you succeed in implementing a neural net on the RCX? I'm sure it IS possible, if not with NQC surely with legOS (though with NQC too we succeeded in emulating (...) (25 years ago, 24-Nov-99, to lugnet.robotics)
|
3 Messages in This Thread:
 
- Entire Thread on One Page:
- Nested:
All | Brief | Compact | Dots
Linear:
All | Brief | Compact
|
|
|
|
