CoG2019-GVGAI Learning Competition

Organised by Hao Tong1, Ruben Rodriguez Torrado2 and Philip Bontrager2

Steering Committee: Jialin Liu1, Julian Togelius2, Diego Perez-Liebana3 and Simon M. Lucas3

1 Southern University of Science and Technology (SUSTech), China
2 New York University (NYU), USA
3 Queen Mary University of London (QMUL), UK

The 3rd GVGAI Single-Player Learning Competition is organised at the IEEE’s 2019 Conference on Games (CoG2019) .

More about the GVGAI competitions can be found here

Please refer to this survey for more about the research using GVGAI framework:

General Video Game AI: a Multi-Track Framework for Evaluating Agents, Games and Content Generation Algorithms. Diego Perez-Liebana, Jialin Liu, Ahmed Khalifa, Raluca D. Gaina, Julian Togelius, Simon M. Lucas.


Due to the long training time, the GVGAI server won’t be used for training your agent. Please train your agent using your own machine or server.


Download and set up the new GVGAI-Gym framework on your machine/server.

Demo video on YouTube

Training phase

Program your agent and train it

  • On as many games/levels as you want
  • Using as much time as you want for deciding an action per game tick
  • Using as much time as you want for training

Validate Phase

  • We will release several levels for some games for training. Unknown levels of the same games will be used for validation.
  • Train your agent on the given levels of given games using as much time as you want.
  • You can download from this webpage. The training games will be embeded to the framework


  • Zip your trained agent.
  • Submit your agent to the competition.
  • Submission will be opened from 1st May 2019 (GMT) to 15th July 2019, 23:59 (GMT) .

Bug Report Phase

  • Submission will be closed at 15th July 2019, 23:59 (GMT).
  • You will be contacted if we have problems running your agent on our server.