Fast Graph Generation via Autoregressive Noisy Filtration Modeling
Fuente:
arXiv
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866908834830745600 |
|---|---|
| author | Krimmel, Markus Wiens, Jenna Borgwardt, Karsten Chen, Dexiong |
| author_facet | Krimmel, Markus Wiens, Jenna Borgwardt, Karsten Chen, Dexiong |
| contents | Existing graph generative models often face a critical trade-off between sample quality and generation speed. We introduce Autoregressive Noisy Filtration Modeling (ANFM), a flexible autoregressive framework that addresses both challenges. ANFM leverages filtration, a concept from topological data analysis, to transform graphs into short sequences of subgraphs. We identify exposure bias as a potential hurdle in autoregressive graph generation and propose noise augmentation and reinforcement learning as effective mitigation strategies, which allow ANFM to learn both edge addition and deletion operations. This unique capability enables ANFM to correct errors during generation by modeling non-monotonic graph sequences. Our results show that ANFM matches state-of-the-art diffusion models in quality while offering over 100 times faster inference, making it a promising approach for high-throughput graph generation. The source code is publicly available at https://github.com/BorgwardtLab/anfm . |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2502_02415 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Fast Graph Generation via Autoregressive Noisy Filtration Modeling Krimmel, Markus Wiens, Jenna Borgwardt, Karsten Chen, Dexiong Machine Learning Existing graph generative models often face a critical trade-off between sample quality and generation speed. We introduce Autoregressive Noisy Filtration Modeling (ANFM), a flexible autoregressive framework that addresses both challenges. ANFM leverages filtration, a concept from topological data analysis, to transform graphs into short sequences of subgraphs. We identify exposure bias as a potential hurdle in autoregressive graph generation and propose noise augmentation and reinforcement learning as effective mitigation strategies, which allow ANFM to learn both edge addition and deletion operations. This unique capability enables ANFM to correct errors during generation by modeling non-monotonic graph sequences. Our results show that ANFM matches state-of-the-art diffusion models in quality while offering over 100 times faster inference, making it a promising approach for high-throughput graph generation. The source code is publicly available at https://github.com/BorgwardtLab/anfm . |
| title | Fast Graph Generation via Autoregressive Noisy Filtration Modeling |
| topic | Machine Learning |
| url | https://arxiv.org/abs/2502.02415 |