changing the render options (render_crosshair/weapon/decals/particles),.using available game variables except for POSITION_X, POSITION_Y, POSITION_Z.using available buttons except for CROUCH,.We also reserve the right to disqualify bots that behave random-like or/and are unintelligent in an evident way or have programed malicious behaviour.ĭuring the contest, ViZDoom will have somewhat limited capabilities. Teams and bots aren’t allowed to cooperate.īefore the actual evaluation, the submitted controller will have to pass a simple test - winning a match against bots from ViZDoom. EvaluationĮach team is allowed to have a single submission with 1 bot for a given point in time - new submissions will override old ones. Singleplayer (track 1) challenge page is available here. Agents will compete in multiplayer games and the best frag collector will emerge victorious. Track 2 is a full on Doom deathmatch (like in previous years) on unknown maps. The agent has to use the ViZDoom framework to connect to the game. The task here is to create an Artificial Intelligence agent that is able to compete with other agents in Doom deathmatches using only data available to regular players without any auxilliary information. This competition is run on Computational Intelligence and Games Conference 2018. Visual Doom AI Competition at CIG2018 Multiplayer Track (2) Winner: bwbell (Ben Bell) 1st Runner-Up: TSAILAB (Tsinghua University & Tencent AI Lab) 2nd Runner-Up: michaelkrax (Michael Krax)
We’ve added two best bots from 2017’s edition and the best bot from 2016’s edition. Resultsįrom 33 teams submitted 152 agents, we’ve selected 3 best submissions from 3 different teams according to leaderboard for final evaluation on 10 new unknown maps. UPDATE: Added results of the final evaluation.
0 kommentar(er)
