BeamVQ: Aligning Space-Time Forecasting Model via Self-training on Physics-aware Metrics
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866910460313337856 |
|---|---|
| author | Wu, Hao Shi, Xingjian Huang, Ziyue Zhao, Penghao Xiong, Wei Xue, Jinbao Tao, Yangyu Huang, Xiaomeng Wang, Weiyan |
| author_facet | Wu, Hao Shi, Xingjian Huang, Ziyue Zhao, Penghao Xiong, Wei Xue, Jinbao Tao, Yangyu Huang, Xiaomeng Wang, Weiyan |
| contents | Data-driven deep learning has emerged as the new paradigm to model complex physical space-time systems. These data-driven methods learn patterns by optimizing statistical metrics and tend to overlook the adherence to physical laws, unlike traditional model-driven numerical methods. Thus, they often generate predictions that are not physically realistic. On the other hand, by sampling a large amount of high quality predictions from a data-driven model, some predictions will be more physically plausible than the others and closer to what will happen in the future. Based on this observation, we propose \emph{Beam search by Vector Quantization} (BeamVQ) to enhance the physical alignment of data-driven space-time forecasting models. The key of BeamVQ is to train model on self-generated samples filtered with physics-aware metrics. To be flexibly support different backbone architectures, BeamVQ leverages a code bank to transform any encoder-decoder model to the continuous state space into discrete codes. Afterwards, it iteratively employs beam search to sample high-quality sequences, retains those with the highest physics-aware scores, and trains model on the new dataset. Comprehensive experiments show that BeamVQ not only gave an average statistical skill score boost for more than 32% for ten backbones on five datasets, but also significantly enhances physics-aware metrics. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2405_17051 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | BeamVQ: Aligning Space-Time Forecasting Model via Self-training on Physics-aware Metrics Wu, Hao Shi, Xingjian Huang, Ziyue Zhao, Penghao Xiong, Wei Xue, Jinbao Tao, Yangyu Huang, Xiaomeng Wang, Weiyan Machine Learning Artificial Intelligence Data-driven deep learning has emerged as the new paradigm to model complex physical space-time systems. These data-driven methods learn patterns by optimizing statistical metrics and tend to overlook the adherence to physical laws, unlike traditional model-driven numerical methods. Thus, they often generate predictions that are not physically realistic. On the other hand, by sampling a large amount of high quality predictions from a data-driven model, some predictions will be more physically plausible than the others and closer to what will happen in the future. Based on this observation, we propose \emph{Beam search by Vector Quantization} (BeamVQ) to enhance the physical alignment of data-driven space-time forecasting models. The key of BeamVQ is to train model on self-generated samples filtered with physics-aware metrics. To be flexibly support different backbone architectures, BeamVQ leverages a code bank to transform any encoder-decoder model to the continuous state space into discrete codes. Afterwards, it iteratively employs beam search to sample high-quality sequences, retains those with the highest physics-aware scores, and trains model on the new dataset. Comprehensive experiments show that BeamVQ not only gave an average statistical skill score boost for more than 32% for ten backbones on five datasets, but also significantly enhances physics-aware metrics. |
| title | BeamVQ: Aligning Space-Time Forecasting Model via Self-training on Physics-aware Metrics |
| topic | Machine Learning Artificial Intelligence |
| url | https://arxiv.org/abs/2405.17051 |