Dota 2

“One of the things that worked actually is during the Dota two days when we were trying to use RL to beat this video, beat people at a video game, we put up like a leaderboard and people could just see how different ideas were performing and what was You know, …”

“We were involved even in 2016 when initially OpenAI started, in fact, Azure was even the first sponsor, I think. And then they were doing a lot more reinforcement learning at that time. I remember the Dota two competition, I think, happened on Azure, and then …”

“And there was real insight that was generated there, which was that it really strengthened our belief that scale was the path to improving artificial intelligence. That with Dota two, the secret idea was that we could take huge amounts of experience and feed i…”

“If you're in India, you want to ensure that the publisher of that game is taking enough interest in India to build around that game. For example, there's a game called Dota two for say, it does not have an active India publisher interest in India. However, it'…”

“The Dota II team, Jakub Pachoki, who was doing this, had developed a A technique that allowed you to put lots of data and compute into solving the problem. And it was really working for Dota. And he came over and started doing this for robotics. And we went f…”

“I think Dota two is the last one.”

“The international for Dota two, they're a big tournament. They are, they do a really good job integrating units bought in the e-sports space in Dota gets a revenue split with the championship, the winner. So they're the largest prize pool in e-sports, not even…”

“Actually the way that we selected the game was we went on, on Twitch and just looked down the list of most popular games in the world and starting, you know, number one is League of Legends. The thing about League of Legends is it doesn't run on Linux and it d…”

“Originally, we were downloading all of them every day, and realized that was about two terabytes worth of data a day. That adds up quite quickly, so we ended up filtering down to the sort of most expert players.”

“So I leave, there is nothing, I come back, there is this reinforcement learning bot, and actually, it's beating our scripted bot after, like a week worth of engineering effort. Possibly it was two weeks, but it was something very miniature compared to the deve…”

“There was a point after the event where we set up this big LAN party where we had, like, 50 computers running the bot. We kind of unleashed this swarm of humans to kind of add our bot. And they find, found all the exploits”

“It was like one of the major examples, one of the major examples of the things that we kind of didn't have explicit incentive for, and yet the bot actually learned them.”

“We first played our TZ who showed up on, on our switch bot, you know, kind of the Franken bot. And you know, that beat him three times.”