EPISODE · Jun 2, 2026 · 7 MIN
How to get better at video games, according to babies - Brian Christian
from TED-Ed · host TED-Ed
Explore what happened when the AI system Deep Q Networks (DQN) attempted to beat the Atari game Montezuma’s Revenge.--In 2013, a group of researchers wanted to create an AI system that could beat every Atari game. They developed a system called Deep Q Networks (DQN) and less than two years later, it was superhuman. But there was one notable exception. When playing Montezuma’s Revenge, DQN couldn’t score a single point. What was it that made this game so vexingly difficult for AI? Brian Christian investigates.Lesson by Brian Christian, directed by Gavin Edwards, Movult.Support Our Non-Profit Mission----------------------------------------------Support us on Patreon: http://bit.ly/TEDEdPatreonCheck out our merch: http://bit.ly/TEDEDShop----------------------------------------------Connect With Us----------------------------------------------Sign up for our newsletter: http://bit.ly/TEDEdNewsletterFollow us on Facebook: http://bit.ly/TEDEdFacebookFind us on Twitter: http://bit.ly/TEDEdTwitterPeep us on Instagram: http://bit.ly/TEDEdInstagram----------------------------------------------Keep Learning----------------------------------------------View full lesson: https://ed.ted.com/lessons/how-to-get-better-at-video-games-according-to-babies-brian-christianDig deeper with additional resources: https://ed.ted.com/lessons/how-to-get-better-at-video-games-according-to-babies-brian-christian#digdeeperCheck out our full book recommendation: https://shop.ed.ted.com/collections/ted-ed-book-recommendations/products/copy-of-1991Animator's website: https://www.movult.com----------------------------------------------Thank you so much to our patrons for your support! Without you this video would not be possible! JY Kang, Abhishek Goel, Heidi Stolt, Christina Salvatore, Karlee Finch, Michael Goldberg, Denise A Pitts, Doug Henry, Keven Webb, Mihai Sandu, Deepak Iyer, Javid Gozalov, Kyanta Yap, Rebecca Reineke, William Biersdorf, Patricia Alves Panagides, Yvette Mocete, Cyrus Garay, Samuel Barbas, Brighteagle, LadyGeek, Curtis Light, Dianne Palomar, Marin Kovachev, Fahad Nasser Chowdhury, Penelope Misquitta, Hans Peng, Gaurav Mathur, Erik Biemans, Tony, Michelle, Katie and Josh Pedretti, Sunny Patel, Hoai Nam Tran, Stina Boberg, Kack-Kyun Kim, Michael Braun-Boghos, zjweele13, Jurjen Geleijn, Anna-Pitschna Kunz, Edla Paniguel, Thomas Mungavan, Jaron Blackburn, Venkat Venkatakrishnan, ReuniteKorea, Aaron Henson, Rohan Gupta, Begum Tutuncu, Mikhail Shkirev, Brian Richards, Cindy O. and Jørgen Østerpart. Learn more about your ad choices. Visit megaphone.fm/adchoices
What this episode covers
Explore what happened when the AI system Deep Q Networks (DQN) attempted to beat the Atari game Montezuma’s Revenge.--In 2013, a group of researchers wanted to create an AI system that could beat every Atari game. They developed a system called Deep Q Networks (DQN) and less than two years later, it was superhuman. But there was one notable exception. When playing Montezuma’s Revenge, DQN couldn’t score a single point. What was it that made this game so vexingly difficult for AI? Brian Christian investigates.Lesson by Brian Christian, directed by Gavin Edwards, Movult.Support Our Non-Profit Mission----------------------------------------------Support us on Patreon: http://bit.ly/TEDEdPatreonCheck out our merch: http://bit.ly/TEDEDShop----------------------------------------------Connect With Us----------------------------------------------Sign up for our newsletter: http://bit.ly/TEDEdNewsletterFollow us on Facebook: http://bit.ly/TEDEdFacebookFind us on Twitter: http://bit.ly/TEDEdTwitterPeep us on Instagram: http://bit.ly/TEDEdInstagram----------------------------------------------Keep Learning----------------------------------------------View full lesson: https://ed.ted.com/lessons/how-to-get-better-at-video-games-according-to-babies-brian-christianDig deeper with additional resources: https://ed.ted.com/lessons/how-to-get-better-at-video-games-according-to-babies-brian-christian#digdeeperCheck out our full book recommendation: https://shop.ed.ted.com/collections/ted-ed-book-recommendations/products/copy-of-1991Animator's website: https://www.movult.com----------------------------------------------Thank you so much to our patrons for your support! Without you this video would not be possible! JY Kang, Abhishek Goel, Heidi Stolt, Christina Salvatore, Karlee Finch, Michael Goldberg, Denise A Pitts, Doug Henry, Keven Webb, Mihai Sandu, Deepak Iyer, Javid Gozalov, Kyanta Yap, Rebecca Reineke, William Biersdorf, Patricia Alves Panagides, Yvette Mocete, Cyrus Garay, Samuel Barbas, Brighteagle, LadyGeek, Curtis Light, Dianne Palomar, Marin Kovachev, Fahad Nasser Chowdhury, Penelope Misquitta, Hans Peng, Gaurav Mathur, Erik Biemans, Tony, Michelle, Katie and Josh Pedretti, Sunny Patel, Hoai Nam Tran, Stina Boberg, Kack-Kyun Kim, Michael Braun-Boghos, zjweele13, Jurjen Geleijn, Anna-Pitschna Kunz, Edla Paniguel, Thomas Mungavan, Jaron Blackburn, Venkat Venkatakrishnan, ReuniteKorea, Aaron Henson, Rohan Gupta, Begum Tutuncu, Mikhail Shkirev, Brian Richards, Cindy O. and Jørgen Østerpart. Learn more about your ad choices. Visit megaphone.fm/adchoices
NOW PLAYING
How to get better at video games, according to babies - Brian Christian
No transcript for this episode yet
Similar Episodes
Apr 13, 2026 ·41m
Apr 13, 2026 ·40m
Apr 13, 2026 ·64m
Apr 13, 2026 ·48m
Apr 9, 2026 ·55m
Apr 1, 2026 ·42m