This podcast is brought to you in part by PNAS Science Sessions, a production of the proceedings

of the National Academy of Sciences. Science Sessions offers brief yet insightful discussions

with some of the world's top researchers. Just in time for the spooky season of Halloween,

we invite you to explore the extraordinary hunting abilities of spiders featuring impressive

aerial maneuvers and webs that function as sensory antennas, follow science sessions,

on popular podcast platforms like iTunes, Spotify, or your preferred podcast platform.

This is Scientific American's 60-Second Science, I'm Pakina Mimer.

Whether you're a pro gamer or you dip your toes into that world every once in a while,

chances are you got stuck while playing a video game once, or was even gloriously defeated by one.

I know I have. Maybe in your frustration, you kicked the console a little. Maybe you took it

out on the controllers, or if you're an 80s kid like me, made the joystick pay.

Now a group of computer scientists from Uber AI are taking revenge for all of us who've been in

this situation before. Using a family of simple algorithms, tact go explore,

they went back and beat some of the most notoriously difficult Atari games,

whose chunky blocks of pixels and 8-bit tunes had once challenged, taunted, and even enraged us.

But what is revisiting those games from the 80s and 90s accomplish,

besides fulfilling a childhood fantasy, according to the scientists who publish their work in nature,

experimenting with solving video games that require a complex, hard exploration gives

way to better learning algorithms. They become more intelligent and perform better under real world

scenarios. One of the nice things of Go Explorer is that it's not just limited to video games,

but that you can also apply to practical applications like robotics.

That was Yoast Hazencha, one of the principal researchers at Uber AI. Yoast developed Go Explorer

...