Deep Surrogate Assisted MAP-Elites for Automated Hearthstone Deckbuilding

12/07/2021
by   Yulun Zhang, et al.
0

We study the problem of efficiently generating high-quality and diverse content in games. Previous work on automated deckbuilding in Hearthstone shows that the quality diversity algorithm MAP-Elites can generate a collection of high-performing decks with diverse strategic gameplay. However, MAP-Elites requires a large number of expensive evaluations to discover a diverse collection of decks. 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 diverse dataset to improve the surrogate model accuracy, while the surrogate model helps guide MAP-Elites towards promising new content. In a Hearthstone deckbuilding case study, we show that our approach improves the sample efficiency of MAP-Elites and outperforms a model trained offline with random decks, as well as a linear surrogate model baseline, setting a new state-of-the-art for quality diversity approaches in automated Hearthstone deckbuilding.

READ FULL TEXT

page 2

page 8

research
06/15/2018

Data-Efficient Design Exploration through Surrogate-Assisted Illumination

Design optimization techniques are often used at the beginning of the de...
research
05/12/2022

Surrogate Infeasible Fitness Acquirement FI-2Pop for Procedural Content Generation

When generating content for video games using procedural content generat...
research
05/07/2021

SimJEB: Simulated Jet Engine Bracket Dataset

This paper introduces the Simulated Jet Engine Bracket Dataset (SimJEB):...
research
05/10/2021

Designing Air Flow with Surrogate-assisted Phenotypic Niching

In complex, expensive optimization domains we often narrowly focus on fi...
research
03/29/2017

Hierarchical Surrogate Modeling for Illumination Algorithms

Evolutionary illumination is a recent technique that allows producing ma...
research
02/13/2017

Data-Efficient Exploration, Optimization, and Modeling of Diverse Designs through Surrogate-Assisted Illumination

The MAP-Elites algorithm produces a set of high-performing solutions tha...
research
03/29/2021

Pairing Character Classes in a Deathmatch Shooter Game via a Deep-Learning Surrogate Model

This paper introduces a surrogate model of gameplay that learns the mapp...

Please sign up or login with your details

Forgot password? Click here to reset