On the Foundations of Earth and Climate Foundation Models
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2024
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866916238113898496 |
|---|---|
| author | Zhu, Xiao Xiang Xiong, Zhitong Wang, Yi Stewart, Adam J. Heidler, Konrad Wang, Yuanyuan Yuan, Zhenghang Dujardin, Thomas Xu, Qingsong Shi, Yilei |
| author_facet | Zhu, Xiao Xiang Xiong, Zhitong Wang, Yi Stewart, Adam J. Heidler, Konrad Wang, Yuanyuan Yuan, Zhenghang Dujardin, Thomas Xu, Qingsong Shi, Yilei |
| contents | Foundation models have enormous potential in advancing Earth and climate sciences, however, current approaches may not be optimal as they focus on a few basic features of a desirable Earth and climate foundation model. Crafting the ideal Earth foundation model, we define eleven features which would allow such a foundation model to be beneficial for any geoscientific downstream application in an environmental- and human-centric manner.We further shed light on the way forward to achieve the ideal model and to evaluate Earth foundation models. What comes after foundation models? Energy efficient adaptation, adversarial defenses, and interpretability are among the emerging directions. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2405_04285 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | On the Foundations of Earth and Climate Foundation Models Zhu, Xiao Xiang Xiong, Zhitong Wang, Yi Stewart, Adam J. Heidler, Konrad Wang, Yuanyuan Yuan, Zhenghang Dujardin, Thomas Xu, Qingsong Shi, Yilei Artificial Intelligence Signal Processing Foundation models have enormous potential in advancing Earth and climate sciences, however, current approaches may not be optimal as they focus on a few basic features of a desirable Earth and climate foundation model. Crafting the ideal Earth foundation model, we define eleven features which would allow such a foundation model to be beneficial for any geoscientific downstream application in an environmental- and human-centric manner.We further shed light on the way forward to achieve the ideal model and to evaluate Earth foundation models. What comes after foundation models? Energy efficient adaptation, adversarial defenses, and interpretability are among the emerging directions. |
| title | On the Foundations of Earth and Climate Foundation Models |
| topic | Artificial Intelligence Signal Processing |
| url | https://arxiv.org/abs/2405.04285 |