Voit lisätä twiitteihisi sijainnin, esimerkiksi kaupungin tai tarkemman paikan, verkosta ja kolmannen osapuolen sovellusten kautta. Halutessasi voit poistaa twiittisi sijaintihistorian myöhemmin. Lue lisää
They say 'using only pixels... as inputs', which implies that they did have the same constraints (except maybe display output latency and controller input latency).
Right. Display and controller latency is the part I'm wondering about. Does it change if the AI is connected to a camera looking at a monitor and can only signal to a controller/kb+m. I didn't find any details about it in the deepmind article.
I suppose an easier test would be to just artificially limit/delay the AI by a given ms. *shrug*
They did say the added an extra 1/4 second delay as part of the test, and agents still outperformed humans.
That would certainly do it. Thanks.
Twitter saattaa olla ruuhkautunut tai ongelma on muuten hetkellinen. Yritä uudelleen tai käy Twitterin tilasivulla saadaksesi lisätietoja.
