Combined Representation and Generation with Diffusive State Predictive Information Bottleneck
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2025
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866909837004111872 |
|---|---|
| author | John, Richard Qiu, Yunrui Herron, Lukas Tiwary, Pratyush |
| author_facet | John, Richard Qiu, Yunrui Herron, Lukas Tiwary, Pratyush |
| contents | Generative modeling becomes increasingly data-intensive in high-dimensional spaces. In molecular science, where data collection is expensive and important events are rare, compression to lower-dimensional manifolds is especially important for various downstream tasks, including generation. We combine a time-lagged information bottleneck designed to characterize molecular important representations and a diffusion model in one joint training objective. The resulting protocol, which we term Diffusive State Predictive Information Bottleneck (D-SPIB), enables the balancing of representation learning and generation aims in one flexible architecture. Additionally, the model is capable of combining temperature information from different molecular simulation trajectories to learn a coherent and useful internal representation of thermodynamics. We benchmark D-SPIB on multiple molecular tasks and showcase its potential for exploring physical conditions outside the training set. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2510_09784 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Combined Representation and Generation with Diffusive State Predictive Information Bottleneck John, Richard Qiu, Yunrui Herron, Lukas Tiwary, Pratyush Machine Learning Statistical Mechanics Quantitative Methods Generative modeling becomes increasingly data-intensive in high-dimensional spaces. In molecular science, where data collection is expensive and important events are rare, compression to lower-dimensional manifolds is especially important for various downstream tasks, including generation. We combine a time-lagged information bottleneck designed to characterize molecular important representations and a diffusion model in one joint training objective. The resulting protocol, which we term Diffusive State Predictive Information Bottleneck (D-SPIB), enables the balancing of representation learning and generation aims in one flexible architecture. Additionally, the model is capable of combining temperature information from different molecular simulation trajectories to learn a coherent and useful internal representation of thermodynamics. We benchmark D-SPIB on multiple molecular tasks and showcase its potential for exploring physical conditions outside the training set. |
| title | Combined Representation and Generation with Diffusive State Predictive Information Bottleneck |
| topic | Machine Learning Statistical Mechanics Quantitative Methods |
| url | https://arxiv.org/abs/2510.09784 |