[tree_of_item]

An artificial intelligence observed a game being played and recreated it and you aren't impressed at all? Man, the future must be boring for people living in it.

Yes, the technique it uses only works for a certain space of possible games. That means there is an obvious path to increasing the size of that space.

[AstralStorm]

"Observed" after being fed lots of sprites and actual ways on how to play it and actually win at it in the objective function. And it "played" only one kind of game. "Obvious path" riiiight.

In addition, this is wrong to having been said to be new, such attempts have been made before and even stronger in results and generality. For example this (relatively dumb) approach from 2013 generalized kinda well, much better than I've seen a silly even deep network generalize: http://www.cs.cmu.edu/~tom7/mario/

So yes, they are overselling it a lot. I am 100% not impressed by this paper as it lacks critical detail. That it can parse stuff from 2D frames is not interesting, it is basic motion analysis which can be done even by a supremely stupid algorithm, not even a CNN.

I mean, Google best AI can play 15 rooms of a simple game...

[elliottcarlson]

You are comparing a system that learned to play a game (which indeed was very impressive), to a system that learned to make the game by observing the behavior from video. None of your points actually relate to the system described.

[AstralStorm]

By "make" you meant "match some sort of a simple function approximation after hardcoding lots of knowledge about the system and the general function" right? Which is essentially what the neural networks and all the other optimization algorithms were made for?

(The algorithm as described will require a huge database for a game that is even slightly more complicated than Infinite Mario. And we don't even have the sources to try that.)

Even the object motion tracker part will choke in 3D environment. (It is a greedy matcher as they described it.)

Speaking of impressed, Google DeepMind paper is way more actually feasible to improve upon and rich in detail: https://arxiv.org/pdf/1606.01868v1.pdf Compare the two papers in straight quality. I understand why you'd publish any worthless junk in the current academic culture and do not agree we should actually do it.

[Bartweiss]

My complaint is that the path to improving their space is "humans hardcoding endless rule lists".

Section 3.1 of the paper outlines a list of 'hand-authored' functions the agent used to derive events from images. They include animation, sprite-entity relationships, motion, collision, and camera movement. Which is to say, every component of Super Mario level 1-1.

That doesn't mean the paper is uninteresting, or useless. Defining facts based on those possible rules is still an intriguing result. I'm having real trouble working out from the paper how well their agent understood conditional changes like size and fire flowers - if it accurately recreated those rules, then I am impressed.

But "modeled without accessing the code" is a dubious claim about an agent that started with a list of the core rules included in its code. The Engine Learning section (3.2) mentions that automatic derivation of possible facts is a key area for future work. That is to say "this would be flexible if it did feature learning instead of needing feature engineering". Unfortunately, that's the problem in agent design, and the value of CNNs isn't unbeatable performance but the capacity for flexible feature learning. The press release here elides the issue of feature learning entirely when comparing performance.