Level Generation for Angry Birds with Sequential VAE and Latent Variable Evolution
This addresses the challenge of automating level generation for physics-based games like Angry Birds, which is incremental as it adapts deep generative models to a specific domain.
The authors tackled the problem of generating playable levels for Angry Birds, a game with non-tile-based representation, by proposing a sequential encoding method that processes levels as text data, resulting in drastically improved stability and diversity compared to existing approaches.
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.