Results of the 2014 Man vs Machine Challenge

We set up 6 computer were people could play. Humans played first from 9am to 4.30pm. 

Every participant had to solve 4 Angry Birds levels in 10 minutes. The levels were very hard and some humans solved only one of them while others gave up. The very best players solved all four levels. 

At 4.30pm the four best AI agents from the 2014 Angry Birds AI Competition were started. We first ran the third and fourth ranked agents and then the second and first ranked agents.  
PlanA+ and DataLab Birds were the two best performing agents of the challenge, but they both solved only two of the four levels. IHSEV solved only one level, while AngryBER didn't manage to solve a level. Therefore:

The winner of the 2014 Man vs Machine Challenge is A HUMAN!!!!

Stefan Ellmauthaler from Leipzig had the highest score of 258,340 and won the 2014 Man vs Machine Challenge. Second place went to Radim Spetlik (251,080), third place to Jan Nykl (242,880), fourth place to Michal Sroka (240,920), and fifth place to Ringo Baumann (240,250). Well done humans!

It is good to see that we are still better at playing Angry Birds than AI and Angry Birds remains a very challenging problem for Artificial Intelligence. PlanA+, the best AI agent of the Challenge was ranked 25th with 168,900 points and Datalab Birds was ranked 29th with 165,580 points, which means they were better than about two thirds of the human players. While they are clearly not as good as the best human players yet, this is already a remarkable achievement!  


A big THANK YOU to all the participants. We hope to see you again in 2015 for an even more exciting Man vs Machine Challenge.

Humans: keep practising

AI Researchers, Developers, Enthusiasts, Hobbyists, Students and anyone else who would like to contribute, you can download our basic game playing software and start adding your own Angry Birds strategies to it or put some AI techniques to the test. Let's see how good we can be in 2015!


Jochen Renz, XiaoYu (Gary) Ge, Peng Zhang and Stephen Gould

Australian National University