Retraction to: Nature https://doi.org/10.1038/nature17620Published online 13 April 2016

We, the authors, are regretfully retracting this Article owing to an error in our computer code that means the quantitative results reported are not valid. We thank A. Grønlund and D. Sels, whose independent efforts1,2 pointed to potential problems with our optimization algorithm. The error was identified by A. Grønlund, who has provided a detailed account3 of the error and its effect on the quantitative results in our Article. For more recent and comprehensive explorations of the performance differences between player-seeded and randomly seeded algorithms, we refer to our recent work4.