Dec 7, 2021 · We propose assisting MAP-Elites with a deep surrogate model trained online to predict game outcomes with respect to candidate decks. MAP-Elites ...

Previous work on automated deckbuilding in Hearthstone shows that the quality diversity algorithm MAP-Elites can generate a collection of high-performing decks ...

The approach improves the sample efficiency of MAP-Elites and outperforms a model trained offline with random decks, as well as a linear surrogate model ...

Jul 9, 2022 · We propose assisting MAP-Elites with a deep surrogate model trained online to predict game outcomes with respect to candidate decks. MAP-Elites ...

This project contains the Hearthstone experiment of the paper Deep Surrogate Assisted MAP-Elites for Automated Hearthstone Deckbuilding. The code base ...

Dec 7, 2021 · We study the problem of efficiently generating high-quality and diverse content in games. Previous work on automated deckbuilding in ...

Deep surrogate assisted MAP-elites for automated hearthstone deckbuilding ... Clune , " Scaling map-elites to deep neuroevolution," in Proceedings of the ...

We propose assisting MAP-Elites with a deep surrogate model trained online to predict game outcomes with respect to candidate decks. MAP-Elites discovers a ...

Co-authors ; Deep surrogate assisted map-elites for automated hearthstone deckbuilding. Y Zhang, MC Fontaine, AK Hoover, S Nikolaidis. Proceedings of the Genetic ...

Apr 19, 2022 · Excited to share a new paper “Deep Surrogate Assisted MAP-Elites for Automated Hearthstone Deckbuilding'' with @tehqin17, @amykhoover, ...