Salvato in:
| Autori principali: | , , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | https://arxiv.org/abs/2509.09919 |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866910000490741760 |
|---|---|
| author | Yiu, Franklin Lu, Mohan Li, Nina Joseph, Kevin Zhang, Tianxu Togelius, Julian Merino, Timothy Earle, Sam |
| author_facet | Yiu, Franklin Lu, Mohan Li, Nina Joseph, Kevin Zhang, Tianxu Togelius, Julian Merino, Timothy Earle, Sam |
| contents | Procedural content generation often requires satisfying both designer-specified objectives and adjacency constraints implicitly imposed by the underlying tile set. To address the challenges of jointly optimizing both constraints and objectives, we reformulate WaveFunctionCollapse (WFC) as a Markov Decision Process (MDP), enabling external optimization algorithms to focus exclusively on objective maximization while leveraging WFC's propagation mechanism to enforce constraint satisfaction. We empirically compare optimizing this MDP to traditional evolutionary approaches that jointly optimize global metrics and local tile placement. Across multiple domains with various difficulties, we find that joint optimization not only struggles as task complexity increases, but consistently underperforms relative to optimization over the WFC-MDP, underscoring the advantages of decoupling local constraint satisfaction from global objective optimization. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2509_09919 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | A Markovian Framing of WaveFunctionCollapse for Procedurally Generating Aesthetically Complex Environments Yiu, Franklin Lu, Mohan Li, Nina Joseph, Kevin Zhang, Tianxu Togelius, Julian Merino, Timothy Earle, Sam Artificial Intelligence Procedural content generation often requires satisfying both designer-specified objectives and adjacency constraints implicitly imposed by the underlying tile set. To address the challenges of jointly optimizing both constraints and objectives, we reformulate WaveFunctionCollapse (WFC) as a Markov Decision Process (MDP), enabling external optimization algorithms to focus exclusively on objective maximization while leveraging WFC's propagation mechanism to enforce constraint satisfaction. We empirically compare optimizing this MDP to traditional evolutionary approaches that jointly optimize global metrics and local tile placement. Across multiple domains with various difficulties, we find that joint optimization not only struggles as task complexity increases, but consistently underperforms relative to optimization over the WFC-MDP, underscoring the advantages of decoupling local constraint satisfaction from global objective optimization. |
| title | A Markovian Framing of WaveFunctionCollapse for Procedurally Generating Aesthetically Complex Environments |
| topic | Artificial Intelligence |
| url | https://arxiv.org/abs/2509.09919 |