Uber AI plays any Atari 2600 game with 'superhuman' skill

DeepMind isn’t the only one with an Atari-savvy AI. A team of Uber AI researchers has developed a set of algorithms, Go-Explore, that reportedly beats any Atari 2600 game with “superhuman” scores, including ones where AI previously had trouble besting its organic rivals. The key is a system that takes care to remember promising states and returns to those states before it sets out exploring.

Go-Explore saw improvement by “orders of magnitude” in some games. It was the first to beat every level in Montezuma’s Revenge, and got a “near-perfect” Pitfall score — both of these are particularly challenging for reinforcement learning systems like this. DeepMind’s Agent57 reached a similar benchmark, according to the team’s Jeff Clune, but through “entirely different methods.” That gives developers a “diversity” of approaches to the same tasks.



from E-Crypto News https://ift.tt/3aR7g9I

Comments

Popular posts from this blog

Market Cipher B — Reversed Engineered and Free

Not ready to go public? That’s the perfect time to think about going public

Oh no!!! The price of BTC is dropping because of this and that. Who cares? I sure don't, because I am betting on the potential of Bitcoin as a medium of exhange in the long term.