Author Topic: A New ANN Learning Algorithm  (Read 893 times)

0 Members and 1 Guest are viewing this topic.

Offline LoligoTopic starter

  • Jr. Member
  • **
  • Posts: 10
  • Helpful? 1
A New ANN Learning Algorithm
« on: September 24, 2010, 02:22:22 PM »
Hi,

Sorry about this folks. I had put up a post here, which got mis-interpreted as a commercial. Also I was mis-characterized as a shyster. To try and offset this impossible-to-overcome impression, and help assure anyone wondering, I'll eliminate all links to my outside contributions and discussions.

For the record my motivations for the original contribution were two-fold:

1. I wanted to notify the community of a new learning algorithm.

2. I wanted to express my passion for the idea of making neural networks more practical for use within robotics, with the hopes of determining if there were any others in the community who share my enthusiasm.

Again, I'm sorry. Anybody who actually DOES share my enthusiasm for making neural networks more usable in antonymous systems (like robots), send me a DM. I'd love to hear from you.

 -L

« Last Edit: September 25, 2010, 06:42:35 PM by Loligo »

Offline knossos

  • Robot Overlord
  • ****
  • Posts: 278
  • Helpful? 14
My interest in AI is what got me into robotics in the first place.  My long term goals include implementing a fuzzy neural network combined with an expert system for my robot.  I'm not sure yet how practical it would be computationally, but thats a long way off still.  For now, being new to robotics, my focus is getting familiarized with the interfacing, capabilities, and limitations of the sensors, processors, etc.  I'm also working on familiarization with mechanical engineering and design.  When I do get to implementing AI, I would be interested in knowing what you think.

Great post by the way.  Avoiding the big elephant, I like this intro to neural networks.
"Never regret thy fall,
O Icarus of the fearless flight
For the greatest tragedy of them all
Is never to feel the burning light."
 
— Oscar Wilde

Offline Soeren

  • Supreme Robot
  • *****
  • Posts: 4,672
  • Helpful? 227
  • Mind Reading: 0.0
So, a commercial for yet another book on AI, this time with the selling point being putting all previous AI's down...

I read your page and it strikes me like the ones selling over unity devices and live-forever-bracelets... Lots of the writers opinions, but not a single line of usable hard facts - if the book you're touting is that good, give away a free chapter to let potential customers judge for themselves and the book will sell itself.

Me... I think I'll wait for the movie  ;)
Regards,
Søren

A rather fast and fairly heavy robot with quite large wheels needs what? A lot of power?
Please remember...
Engineering is based on numbers - not adjectives

Offline LoligoTopic starter

  • Jr. Member
  • **
  • Posts: 10
  • Helpful? 1
Re: A New ANN Learning Algorithm
« Reply #3 on: September 25, 2010, 06:50:04 PM »
Soeren and  knossos,

Hello,

For obvious reasons I'll respond to you privately via PMs when time permits. Apologies to everyone else for the confusion.

-L

Offline z.s.tar.gz

  • Supreme Robot
  • *****
  • Posts: 540
  • Helpful? 5
  • Linux Guru
Re: A New ANN Learning Algorithm
« Reply #4 on: September 27, 2010, 03:00:23 PM »
Your goal to alert the community of a new learning algorithm is in vain. Nobody is browsing around the forums and says "wow this is a really interesting new AI program. I think I'll stop whatever project I'm working on now and start working on some learning algorithms."

If I want to make a learning robot right now, I'll do my research on the googlenet, not by skimming through the forums. If I don't want to make a learning robot, I will disregard the thread entirely as it does not pertain to me.

So in the end it's like car commercials on the radio. No matter how much they tell me about these amazing deals, I'm just not looking for a new car. When it is time for a new car however, I don't turn to the radio for my options. I read up on the internet or the local paper to see who really has the best deal. So either way, I don't pay attention to the car commercials. Or this thread.
Save yourself the typing. Just call me Zach.

 


Get Your Ad Here

data_list