Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Breakout Example #15

Open
kmader opened this issue Mar 22, 2016 · 5 comments
Open

Breakout Example #15

kmader opened this issue Mar 22, 2016 · 5 comments

Comments

@kmader
Copy link

kmader commented Mar 22, 2016

In case it's interesting or you have a sample gallery, breakout with deep-q learning (since that's the youtube video example originally shown for such games)
http://4quant.com/javascript-breakout/ repo: https://github.com/4Quant/javascript-breakout

@gitricko
Copy link

Yes, second that... I probably, will take a stab at that to see if we can integrate ReinforceJS to this js-breakout game.

Another question: Does the javascript version able to leverage the GPU ?

@gitricko
Copy link

gitricko commented Apr 24, 2016

@kmader Just got a simple breakout game working using a simple code from MDN

Actually, before trying out breakout, i was trying reinforcejs on "snake eat apple game" and got some success. Thinking it is easy, i move on to breakout but never got it work properly. I keep changing the game settings like using smoothing paddle movement, adding more attributes to the brain. But in fact, it turns out that the game worse when i pass in more variables.

It turns out that lowering the alpha from 0.005 to 0.0001 does the trick !!! Check it out. You can save your model as well as loading the prebuilding model by click on "Load Model button"

The code is not that clean, abit messy because i was trying out stuff. Will refactor when i have time. This is cool !

Try it ! Note, you can use developer console to check real-time between consecutive hits vs misses.

@gitricko
Copy link

gitricko commented Jun 5, 2016

Got javascript-breakout to work... Check it out !

@alperenkirimer
Copy link

@gitricko Hi we are also working on a deep q learning game using reinforcejs. Just wondering how many iterations do we need to see the ai acting clever?

@gitricko
Copy link

gitricko commented Jun 3, 2019

@alperenkirimer For me, it took a few hours because my environment is simplistics... It may take a day using CPU

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

3 participants