Level generation for angry birds with sequential VAE and latent variable evolution

Abstract

Video game level generation based on machine learning (ML), in particular, deep generative models, has attracted attention as a technique to automate level generation. However, applications of existing ML-based level generations are mostly limited to tile-based level representation. When ML techniques are applied to game domains with non-tile-based level representation, such as Angry Birds, where objects in a level are specified by real-valued parameters, ML often fails to generate playable levels. In this study, we develop a deep-generative-model-based level generation for the game domain of Angry Birds. To overcome these drawbacks, we propose a sequential encoding of a level and process it as text data, whereas existing approaches employ a tile-based encoding and process it as an image. Experiments show that the proposed level generator drastically improves the stability and diversity of generated levels compared with existing approaches. We apply latent variable evolution with the proposed generator to control the feature of a generated level computed through an AI agent's play, while keeping the level stable and natural. CCS CONCEPTS• Computing methodologies → Neural networks; Learning latent representations;

Adaptive scenario subset selection for worst-case optimization and its application to well placement optimization

“…2 That is, the users of this approach need not tune its hyperparameters on their own tasks, but rather can use it out-of-the-box. This property and its superior performance have attracted practitioners, and hence, the CMA-ES has been widely applied to real-world applications [1,2,3,4,5,6,7,8,9,10,11,12,13,14].…”

“…Evolutionary approaches have been successfully applied to simulation-based optimization in a variety of engineering fields. Many examples of such applications have been reported in the relevant literature [1,2,3,4,5,6,7,8,9,10,11,12,13,14]. Evolutionary approaches are preferred for several reasons.…”

ChatGPT4PCG Competition: Character-like Level Generation for Science Birds

“…To create game levels with more stability and variety, a technique called deep generative model-based level generation was used in one study [30]. This involves encoding the level sequentially as text data, which allows for control of level features through latent variable evolution.…”

Covariance Matrix Adaptation Evolutionary Strategy with Worst-Case Ranking Approximation for Min--Max Optimization and its Application to Berthing Control Tasks

“…The proposed approach tries to solve the outer minimization problem of (1) using the CMA-ES. The CMA-ES is a state-of-the-art derivative-free optimization approach for continuous black-box optimization problems [Hansen 2009;Rios and Sahinidis 2013] and has been used in several real-world applications [Fujii et al 2018;Maki et al 2020;Miyagi et al 2018;Tanabe et al 2021;Urieli et al 2011]. There are two essential characteristics of the CMA-ES that attract attention.…”