OpenAI Bots Crush the Best Human Dota 2 Players in the World


After OpenAI took down members of the audience easily during a warmup round, the bot network started and finished the first of the two matches strongly, stopping its human opponents from destroying any of its defensive towers. Today, those bots are playing a benchmark match against five top players, including former pros, casters and analysts.

The firm's Dota 2 team of bots - OpenAI Five - took on semi-professional players ranked in the 99.95th percentile in the world, and beat them 2-1 in a set of three games.

Using a massively scaled-up version of "Proximal Policy Optimization", which is a training system designed for OpenAI's successful 1v1 Dota 2 bot. OpenAI Five plays the game for a whopping 180 years every single day thanks to 256 GPUs and 128,000 CPU cores in the Google Compute Cloud. OpenAI had the capacity to predict its chance of winning during each stage of the game, as well as during character selection. Without the ability to build a team it knew how to play, the AI was at a disadvantage. "This reaction time is much closer to human level, though we haven't seen evidence of changes in gameplay as OpenAI Five's strength comes more from teamwork and coordination than reflexes".

While OpenAI's defeat of some of Dota 2's cream of the crop players is decidedly impressive, what's definitely more noteworthy is the unprecedented advancement of artificial intelligence itself.

More news: Ruby Rose cast as The CW’s Batwoman
More news: Woman helps man short on cash, finds out he's Keith Urban
More news: Blast kills Syrian researcher suspected of links to chemical weapons

The humans won the final game, but only after the audience was allowed to select the bots' player characters, which put them at a significant advantage.

The exhibition match was streamed live on Twitch and featured noted Dota 2 players Blitz, Cap, Fogged, Merlini, and MoonMeander.

We previously reported on the massively complex undertaking it took for OpenAI to train the bots through reinforcement learning, which involved putting them into a virtual world and using trial and error to help them figure out for themselves how to reach a goal.

After losing to OpenAI Five, Capitalist Walsh expressed his despair in a tweet. Well, the AI platform developed by OpenAI would like to have a word, humans.