Jump to content
  • Advertisement

Archived

This topic is now archived and is closed to further replies.

snyp

help with machine learning

This topic is 5359 days old which is more than the 365 day threshold we allow for new replies. Please post a new topic.

If you intended to correct an error in the post then please contact us.

Recommended Posts

were do i begin with artificial intelligence. especially with machine learning..

Share this post


Link to post
Share on other sites
Advertisement
Guest Anonymous Poster
Most machine learning involves a single equation which is easily expressed in most any computer language:

y(t+1) = y(t) + theta * (f(t) - y(t))

in english: with a sufficiently small theta, as more and more samples are taken from f(), y will approximate the mean of the distribution of f().

in practice: Start with a large theta (say 0.5) for fast convergence and gradualy reduce it towards 0 as more and more samples are taken.

Its the basis of the most common learning rule for neural networks, the driving principle behind temporal difference learning, and is suitable for just about any reinforcement learning task.

A good online book for reinforcement learning is:

http://www-anw.cs.umass.edu/~rich/book/the-book.html

- Rockoon

Share this post


Link to post
Share on other sites
quote:
Original post by snyp
were do i begin with artificial intelligence. especially with machine learning..


It may help to get a handle on the various machine learning algorithms which are available. I will suggest reading my article, "Modeling Methodology 3: Algorithm Selection" (PC AI Mar/Apr, 1998) and maybe "Modeling Methodology 4: Localizing Global Models" (PC AI May/Jun, 1998) to get some idea as to what distinguishes various techniques from one another. Both articles are available at:

http://will.dwinnell.com/will/willTechnicalPublications.html

Also, the Usenet comp.ai.neural-nets FAQ may be helpful. See:
http://www.faqs.org/faqs/ai-faq/neural-nets/part1/

-Predictor
http://will.dwinnell.com


Share this post


Link to post
Share on other sites

  • Advertisement
×

Important Information

By using GameDev.net, you agree to our community Guidelines, Terms of Use, and Privacy Policy.

GameDev.net is your game development community. Create an account for your GameDev Portfolio and participate in the largest developer community in the games industry.

Sign me up!