Searching the Latent Space of a Generative Adversarial Network to Generate DOOM Levels

Giacomello, Edoardo; Lanzi, Pier Luca; Loiacono, Daniele

doi:10.1109/cig.2019.8848011

Cited by 21 publications

(18 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…An advantage of using DGMs for level generation is that LVE [7,27] can be employed after training the generator to identity levels with certain characteristics. However, existing studies on sequential VAE have reported that the latent vector tends to be ignored, and the variation of outputs is created by the randomness of G itself [20].…”

Section: Latent Variable Evolutionmentioning

confidence: 99%

“…PCGML includes approaches that utilize n-gram [4] with a graphical probability model [9], generative adversarial networks (GAN) [7,26,27], and variational autoencoders (VAEs) [25]. Volz et al [27] applied the Wasserstein GAN (WGAN) to generate levels for Super Mario Bros (SMB).…”

Section: Introductionmentioning

confidence: 99%

“…An advantage of PCGML using a deep generative models (DGMs), such as GAN or VAE, is the possibility of controlling the characteristics of the generated game contents by applying latent variable evolution (LVE) [7,27]. In these approaches, the game content generator is modeled as a map from a real-valued latent vector to a game content, and is trained on an existing dataset.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

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

Tanabe

Fukuchi

Sakuma

et al. 2021

Proceedings of the Genetic and Evolutionary Computation Conference

View full text Add to dashboard Cite

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;

show abstract

Section: Latent Variable Evolutionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

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

Tanabe

Fukuchi

Sakuma

et al. 2021

Proceedings of the Genetic and Evolutionary Computation Conference

View full text Add to dashboard Cite

show abstract

“…They have also been employed to learn an encoding during optimization, using them as a variational operator [9]. Other GMs like GANs have been used in latent variable evolution [2] to generate levels for the video games Super Mario Bros. [23] and Doom [10]. A model's latent space is searched with an evolutionary algorithm for instances that optimize for desired properties such as the layout or difficulty of a level.…”

Section: Variational Autoencodersmentioning

confidence: 99%

Expressivity of parameterized and data-driven representations in quality diversity search

Hagg

Berns

Asteroth

et al. 2021

Proceedings of the Genetic and Evolutionary Computation Conference

View full text Add to dashboard Cite

We consider multi-solution optimization and generative models for the generation of diverse artifacts and the discovery of novel solutions. In cases where the domain's factors of variation are unknown or too complex to encode manually, generative models can provide a learned latent space to approximate these factors. When used as a search space, however, the range and diversity of possible outputs are limited to the expressivity and generative capabilities of the learned model. We compare the output diversity of a quality diversity evolutionary search performed in two different search spaces: 1) a predefined parameterized space and 2) the latent space of a variational autoencoder model. We find that the search on an explicit parametric encoding creates more diverse artifact sets than searching the latent space. A learned model is better at interpolating between known data points than at extrapolating or expanding towards unseen examples. We recommend using a generative model's latent space primarily to measure similarity between artifacts rather than for search and generation. Whenever a parametric encoding is obtainable, it should be preferred over a learned representation as it produces a higher diversity of solutions.

show abstract

“…An emerging PCG technique is Generative Adversarial Networks (GANs [2]) used to search the latent design space of video game levels, as has been done in Super Mario Bros. [3], Doom [4], an educational game [5], and the General Video Game AI (GVG-AI [6]) adaptation of The Legend of Zelda [7]. In the GVG-AI version of Zelda, single-room levels require the player to fight enemies, reach a key, and take it to the exit.…”

Section: Introductionmentioning

confidence: 99%

Generative Adversarial Network Rooms in Generative Graph Grammar Dungeons for The Legend of Zelda

Gutierrez¹,

Schrum²

2020

Preprint

View full text Add to dashboard Cite

Generative Adversarial Networks (GANs) have demonstrated their ability to learn patterns in data and produce new exemplars similar to, but different from, their training set in several domains, including video games. However, GANs have a fixed output size, so creating levels of arbitrary size for a dungeon crawling game is difficult. GANs also have trouble encoding semantic requirements that make levels interesting and playable. This paper combines a GAN approach to generating individual rooms with a graph grammar approach to combining rooms into a dungeon. The GAN captures design principles of individual rooms, but the graph grammar organizes rooms into a global layout with a sequence of obstacles determined by a designer. Room data from The Legend of Zelda is used to train the GAN. This approach is validated by a user study, showing that GAN dungeons are as enjoyable to play as a level from the original game, and levels generated with a graph grammar alone. However, GAN dungeons have rooms considered more complex, and plain graph grammar's dungeons are considered least complex and challenging. Only the GAN approach creates an extensive supply of both layouts and rooms, where rooms span across the spectrum of those seen in the training set to new creations merging design principles from multiple rooms.

show abstract

Searching the Latent Space of a Generative Adversarial Network to Generate DOOM Levels

Cited by 21 publications

References 21 publications

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

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

Expressivity of parameterized and data-driven representations in quality diversity search

Generative Adversarial Network Rooms in Generative Graph Grammar Dungeons for The Legend of Zelda

Contact Info

Product

Resources

About