This is what makes reinforcement learning research using Doom pretty exciting: if best AI agents we can make lose to humans even in this relatively simple game, there must be something we're missing (whether it's temporal/spatial awareness, memory, or something else) Despite the best efforts of multiple teams none of the participating agents were able to match a human player.
In the last few years (`16-`18), even an AI competition has been held where bots competed against each other in FFA deathmatch.
Some of you might know that Doom is a fairly popular platform amongst us thanks to student doing research in deep reinforcement learning.
0 kommentar(er)
