ResearchBib Share Your Research, Maximize Your Social Impacts
Sign for Notice Everyday Sign up >> Login

2017 - First General Video Game AI Single-Player Learning Competition

Date2017-08-22 - 2017-08-25

Deadline2017-08-06

VenueNew York, USA - United States USA - United States

Keywords

Websitehttp://www.cig2017.com

Topics/Call fo Papers

The deadline of the first General Video Game AI Single-Player Learning Competition is extended to 6th August 2017.
The competition will be organised at the 2017 IEEE Conference on Computational Intelligence in Games (IEEE CIG2017), New York, USA.
Important dates
Submission open: To be announced
Submission deadline: 6 August, 2017
Result announcement: during the conference
Conference: 22-25 August, 2017
Framework
The code for this track is integrated in the master branch of the repository, you can find it here.
The instructions can be found here. It presents the structure of the code, the competition rules, and how to create and run an agent.
- It supports agents written in Java or Python. The agents will be able to select to receive a serialised game state as a string or a screenshot of the game screen.
- The communication is using sockets. The time used by the agents to provide actions, process results and reason about the game in general is measured in CPU time. At the moment this is set to 5 minutes training per game (in three levels), and the time spent on communication is ignored for computing this.
Prizes
- The final ranking will be announced during the conference and will be included in the newsletter of IEEE CIS.
- The IEEE Computational Intelligence Society (IEEE CIS) will delivery official certificates to the winner and the runner-up.
- The IEEE CIS will also award the top-3 student or young professional winners prizes (500USD/300USD/200USD). For the detailed awarding policy, please refer to the webpage of the IEEE CIS Student Game-Based Competitions Sub-Committee.
Thanks for your passion for learning track and patience during the past years. For questions and comments about this track, please contact Jialin Liu (jialin.liu.cn-AT-gmail.com), which is me :)
All your feedback is highly appreciated.
Best wishes and best of luck,
Jialin, Diego, Simon & the GVGAI team http://www.gvgai.net/about_us.php
PS: We are moving to the Queen Mary, University of London from 1st August 2017. For any query to our research team via emails, please include our gmail addresses. Thank you!

Last modified: 2017-07-19 06:38:40