ViZDoom Reinforcement Learning
Visual Doom AI Competition at CIG2018 track 2
This competition is run on Computational Intelligence and Games Conference 2018
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. The agent has to use the ViZDoom framework to connect to the game.
Track 2 is a full on Doom deathmatch (like in previous years) on unknown maps. Agents will compete in multiplayer games and the best frag collector will emerge victorious.
The task is to create bots that fight against each other in a regular deathmatch, where different weapons and items are available. The bots will be ranked by the number of frags, where the number of frags for this competition is defined as: frags = number of killed opponents - number of suicides
During the public evaluations phase bots will play multiple matches with different opponents and the leaderboard will be updated accordingly.
During final 2 weeks the leaderboard will be hidden. Multiple matches on various maps will be played similarly to the public evaluation period. Best bots (probably 8 of them) will take part in a final matches that will determine the top ranking. Finalists will most likely be published before publication of final results at CIG
Each team is allowed to have a single submission with 1 bot for a given point in time - new submissions will override old ones. Teams and bots aren’t allowed to cooperate.
Before the actual evaluation, the submitted controller will have to pass a simple test - winning a match against bots from ViZDoom.
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.
During the contest, ViZDoom will have somewhat limited capabilities.
- loading your config file,
- using ANY resolution,
- using available buttons except for CROUCH,
- using available game variables except for POSITION_X, POSITION_Y, POSITION_Z
- using ANY available screen format,
- changing the render options (render_crosshair/weapon/decals/particles),
- setting agent’s name and color via add_game_args(“+name AI +colorset 0”) (or in a config file),
- network communication other than the game connection itself (unless you need localhost for some reason),
- using new_episode, send_game_command methods (e.g. adding bots)
TECHNICALLY ALLOWED BUT POINTLESS.
- using depth, automap and labels buffers (they will be filled with random noise),
- changing mode, scenario path, vizdoom path,
- using USERX variables,
- using CROUCH button (will not have any effect)
- using POSITION_X, POSITION_Y, POSITION_Z game variables (zeros or noise)
We are looking for sponsors!
Want to help? Contact us!
Repository with a sample (random) submission
- Gitter Channel : crowdAI/vizdoom2018
- Technical Issues : https://github.com/crowdAI/vizdoom2018-singleplayer-starter-kit/issues
- Discussion Forum : https://www.crowdai.org/challenges/visual-doom-ai-competition-2018-track-1/topics
We strongly encourage you to use the public channels mentioned above for communications between the participants and the organizers. In extreme cases, if there are any queries or comments that you would like to make using a private communication channel, then you can send us an email at :