Learning Machine Learning for Game Development
ai.googleblog.com
Learning Machine Learning for Game Development
1–10 of 69 posts
Re: Learning Machine Learning for Game Development
#2You probably mistyped the title. The original title is: LEVERAGING Machine Learning for Game Development.
Re: Learning Machine Learning for Game Development
#3I'm in this research area, it's facinating, but also (like many things in AI) easy to get wrong.
The trickiest bit is deciding exactly what you want your AI to optimise for -- If you want a game where "anyone can win", you end up with coin toss. If you want a game where "skill is everything", you can accidentally end up with "who can click fastest".
On a project I recently worked on (and I'm writing up), we ended up on our 6th or 7th "optimisation function" before we found something where the "AI optimal" seemed actually enjoyable to play.
Of course, if you already have a fairly fixed game, it can be easier to optimise constants, as you already know your target -- but people have also been doing that for a long time previously.
Re: Learning Machine Learning for Game Development
#4What if instead of spending time to train the ML model they just made a dummy client with it’s own simple probabilistic state machine or behaviour tree to balance the game?
How much time and resources would be spent for above mentioned approach compared to ML approach?
I think due to statistical nature of ML, it is seen as kind of an hammer to every problem that might be solved statistically(which there are lots of them) but it might not be the most effective use of engineering resources.
Re: Learning Machine Learning for Game Development
#5This is super cool, but like the other deepmind projects it's a bit frustrating. I've got nothing against "we did something hard with AI and it worked well", but would greatly prefer it to be followed with "... and we're making this available as something you can use too!"
Re: Learning Machine Learning for Game Development
#6What if instead of spending time to train the ML model they just made a dummy client with it’s own simple probabilistic state machine or behaviour tree to balance the game? How much time and resources would be spent for above mentioned approach compared to ML approach? I think due to statistical nature of ML, it is seen as kind of an hammer to every problem that might be solved statistically(which there are lots of t…
Anecdotally: I've never seen a "simple state machine" or "simple behaviour tree" in game AI. We've recently started using deep reinforcement learning for our games and its almost like a miracle how simple, effective and scalable the system is. There are some mentioned problems like designing rewards for player enjoyment, but its definitely got a massive reduction in engineering effort.
Re: Learning Machine Learning for Game Development
#7I got a introduction to ML back in college (going through basic Algorithms) and I went straight to use them in game dev. I think it's a fascinating area.
Re: Learning Machine Learning for Game Development
#8This is super cool, but like the other deepmind projects it's a bit frustrating. I've got nothing against "we did something hard with AI and it worked well", but would greatly prefer it to be followed with "... and we're making this available as something you can use too!"
It's not a DeepMind project. The authors are on the Stadia team.
Re: Learning Machine Learning for Game Development
#9If anyone is having problems loading the page, noscript/disabling javascript will prevent it from displaying a white-only screen.
Re: Learning Machine Learning for Game Development
#10I'm in this research area, it's facinating, but also (like many things in AI) easy to get wrong. The trickiest bit is deciding exactly what you want your AI to optimise for -- If you want a game where "anyone can win", you end up with coin toss. If you want a game where "skill is everything", you can accidentally end up with "who can click fastest". On a project I recently worked on (and I'm writing up), we ended up…
Interesting! Is there some kind of meta-objective for "AI optimal" that could have replaced the 6 or 7 iterations you did with human R&D? For instance, if you had real human playtesters interacting with the prototypes, is there some signal you could extract to measure that it's "good"?