Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2410.24012 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866929570313142272 |
|---|---|
| author | Mercatali, Giangiacomo Verma, Yogesh Freitas, Andre Garg, Vikas |
| author_facet | Mercatali, Giangiacomo Verma, Yogesh Freitas, Andre Garg, Vikas |
| contents | We introduce a novel score-based diffusion framework named Twigs that incorporates multiple co-evolving flows for enriching conditional generation tasks. Specifically, a central or trunk diffusion process is associated with a primary variable (e.g., graph structure), and additional offshoot or stem processes are dedicated to dependent variables (e.g., graph properties or labels). A new strategy, which we call loop guidance, effectively orchestrates the flow of information between the trunk and the stem processes during sampling. This approach allows us to uncover intricate interactions and dependencies, and unlock new generative capabilities. We provide extensive experiments to demonstrate strong performance gains of the proposed method over contemporary baselines in the context of conditional graph generation, underscoring the potential of Twigs in challenging generative tasks such as inverse molecular design and molecular optimization. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2410_24012 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Diffusion Twigs with Loop Guidance for Conditional Graph Generation Mercatali, Giangiacomo Verma, Yogesh Freitas, Andre Garg, Vikas Machine Learning We introduce a novel score-based diffusion framework named Twigs that incorporates multiple co-evolving flows for enriching conditional generation tasks. Specifically, a central or trunk diffusion process is associated with a primary variable (e.g., graph structure), and additional offshoot or stem processes are dedicated to dependent variables (e.g., graph properties or labels). A new strategy, which we call loop guidance, effectively orchestrates the flow of information between the trunk and the stem processes during sampling. This approach allows us to uncover intricate interactions and dependencies, and unlock new generative capabilities. We provide extensive experiments to demonstrate strong performance gains of the proposed method over contemporary baselines in the context of conditional graph generation, underscoring the potential of Twigs in challenging generative tasks such as inverse molecular design and molecular optimization. |
| title | Diffusion Twigs with Loop Guidance for Conditional Graph Generation |
| topic | Machine Learning |
| url | https://arxiv.org/abs/2410.24012 |