Slashdot Mirror
Video Games Are So Realistic That They Can Teach AI What the World Looks Like (vice.com)
Jordan Pearson, reporting for Motherboard:Thanks to the modern gaming industry, we can now spend our evenings wandering around photorealistic game worlds, like the post-apocalyptic Boston of Fallout 4 or Grand Theft Auto V's Los Santos, instead of doing things like "seeing people" and "engaging in human interaction of any kind." Games these days are so realistic, in fact, that artificial intelligence researchers are using them to teach computers how to recognize objects in real life. Not only that, but commercial video games could kick artificial intelligence research into high gear by dramatically lessening the time and money required to train AI. "If you go back to the original Doom, the walls all look exactly the same and it's very easy to predict what a wall looks like, given that data," said Mark Schmidt, a computer science professor at the University of British Columbia (UBC). "But if you go into the real world, where every wall looks different, it might not work anymore." Schmidt works with machine learning, a technique that allows computers to "train" on a large set of labelled data -- photographs of streets, for example -- so that when let loose in the real world, they can recognize, or "predict," what they're looking at. Schmidt and Alireza Shafaei, a PhD student at UBC, recently studied Grand Theft Auto V and found that self-learning software trained on images from the game performed just as well, and in some cases even better, than software trained on real photos from publicly available datasets.
Religious nuts? That is what you AI and Space Nutters are. You think that somehow AI is going to magically appear because your PC got a lot faster between 1990 and 2016.
We tried neural nets back then. Didn't work.
It seems it is you that are stuck in 1960, because connectionist techniques nowadays are nothing like that. And deep-learning has been breaking record after record, even achieving superhuman performance in some tasks. And deep-learning is just a smart math trick over the regular backprop algorithm, allowing more layers without degrading the error gradients. When the models that actually incorporate neuroscientific knowledge (current research) mature, expect even better performances.
And wtf, this "I tried once, I failed, I'll never try again" is surely a loser-talk.
`echo $[0x853204FA81]|tr 0-9 ionbsdeaml`@gmail.com