What does this means?
Author
Hi everyone,
Write and use a smart agent which trains/learns and becomes more accurate or otherwise somehow improves over time
--> what does this means?
Wow, does it means an AI agent, that uses neural networks..
well can anyone give me some examples of agents used in business application today?
Any help, thanks.
Why does it have to be an artificial neural network? There are many methods for learning, many of which are easier to implement and debug than an ANN.
Timkin
Timkin
quote:Original post by Timkin
Why does it have to be an artificial neural network? There are many methods for learning, many of which are easier to implement and debug than an ANN.
Timkin
Just SAYING Neural Network makes some people all warm and squishy inside. People are way too fascinated with GAs, NNs, etc.
Dave Mark - President and Lead Designer
Intrinsic Algorithm - "Reducing the world to mathematical equations!"
[sarcasm]
When I was a kid we played ''guess a number between 1 and 100''
It worked like this:
is it larger than 50 ?
yes
Is it larger than 75 ?
no
is it larger than 63?
yes
etc...
it was easily solved using the ''interval halving'' algorithm.
OMFG!! this solves your problem!
Too bad you can''t use it, because it hasn''t got a kewl name like ''Genetic algorithms'' or ''Neural Networks'' so you won''t impress the other kids...
[/sarcasm]
I always laugh at the people who wants to use a specific algorithm just because it''s got a name that''s sounds like something from a bad sci-fi movie.
HA HAH AHAH AHA HA
Grow up kiddies....
When I was a kid we played ''guess a number between 1 and 100''
It worked like this:
is it larger than 50 ?
yes
Is it larger than 75 ?
no
is it larger than 63?
yes
etc...
it was easily solved using the ''interval halving'' algorithm.
quote:
Write and use a smart agent which trains/learns and becomes more accurate or otherwise somehow improves over time
OMFG!! this solves your problem!
Too bad you can''t use it, because it hasn''t got a kewl name like ''Genetic algorithms'' or ''Neural Networks'' so you won''t impress the other kids...
[/sarcasm]
I always laugh at the people who wants to use a specific algorithm just because it''s got a name that''s sounds like something from a bad sci-fi movie.
HA HAH AHAH AHA HA
Grow up kiddies....
What''s wrong with my post?
It''s sarcastic about the fact that a lot of the people on this forums insists on using NN and GA regardless of the their problem.
When it comes to real-world applications, in use today, in commercially working systems - NN and GA plays such a small role that it''s hardly worth mentioning, yet everybody wants to use the methods because they''ve got cool names.
Remember, useful AI system based on these ''cool'' algorithms has been "right around the corner" for 10-20 yrs, whereas the ''boring'' technologies (search algorithms etc.) has been useful all along...
I usually don''t flame, rant or troll - but the focus on NN and GA is totally out of proportion IMHO.
It''s sarcastic about the fact that a lot of the people on this forums insists on using NN and GA regardless of the their problem.
When it comes to real-world applications, in use today, in commercially working systems - NN and GA plays such a small role that it''s hardly worth mentioning, yet everybody wants to use the methods because they''ve got cool names.
Remember, useful AI system based on these ''cool'' algorithms has been "right around the corner" for 10-20 yrs, whereas the ''boring'' technologies (search algorithms etc.) has been useful all along...
I usually don''t flame, rant or troll - but the focus on NN and GA is totally out of proportion IMHO.
With my last post, I tried to show that a problem should be solved with the simplest possible algorithm - and that most newbies fail to see simple solutions because of their focus on advanced, cool sounding algorithms. Therefore, I think my posts are justified.
well instead of arguing about NN and GA and how they may not be applicable in given situations,or how one may use them simply for the cool sounds factor; we could try helping the poster figure out what he posted about 
This topic is closed to new replies.
Advertisement
Popular Topics
Advertisement
