# Uncertainty and Generalizability in Foundation Models for Earth Observation

arXiv: https://arxiv.org/abs/2409.08744 · HTML: https://mlsanity.com/paper/2409.08744/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 11 of its 13 references that are in the map.
- First posted on arXiv 13 Sep 2024.

## Builds on (most cited first, 11 of 11)

- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [Masked Autoencoders Are Scalable Vision Learners](https://mlsanity.com/paper/2111.06377/) (2021)
- [PointNet++: Deep Hierarchical Feature Learning on Point Sets in a Metric Space](https://mlsanity.com/paper/1706.02413/) (2017)
- [Aleatoric and Epistemic Uncertainty in Machine Learning: An Introduction to Concepts and Methods](https://mlsanity.com/paper/1910.09457/) (2019)
- [SpectralGPT: Spectral Remote Sensing Foundation Model](https://mlsanity.com/paper/2311.07113/) (2023)
- [SkySense: A Multi-Modal Remote Sensing Foundation Model Towards Universal Interpretation for Earth Observation Imagery](https://mlsanity.com/paper/2312.10115/) (2023)
- [Self-supervised Learning in Remote Sensing: A Review](https://mlsanity.com/paper/2206.13188/) (2022)
- [On the Opportunities and Challenges of Foundation Models for Geospatial Artificial Intelligence](https://mlsanity.com/paper/2304.06798/) (2023)
- [SSL4EO-L: Datasets and Foundation Models for Landsat Imagery](https://mlsanity.com/paper/2306.09424/) (2023)
- [Large Scale Masked Autoencoding for Reducing Label Requirements on SAR Data](https://mlsanity.com/paper/2310.00826/) (2023)
- [Fewshot learning on global multimodal embeddings for earth observation tasks](https://mlsanity.com/paper/2310.00119/) (2023)

## Abstract (arXiv)

We take the perspective in which we want to design a downstream task (such as estimating vegetation coverage) on a certain area of interest (AOI) with a limited labeling budget. By leveraging an existing Foundation Model (FM) we must decide whether we train a downstream model on a different but label-rich AOI hoping it generalizes to our AOI, or we split labels in our AOI for training and validating. In either case, we face choices concerning what FM to use, how to sample our AOI for labeling, etc. which affect both the performance and uncertainty of the results. In this work, we perform a large ablative study using eight existing FMs on either Sentinel 1 or Sentinel 2 as input data, and the classes from the ESA World Cover product as downstream tasks across eleven AOIs. We do repeated sampling and training, resulting in an ablation of some 500K simple linear regression models. Our results show both the limits of spatial generalizability across AOIs and the power of FMs where we are able to get over 0.9 correlation coefficient between predictions and targets on different chip level predictive tasks. And still, performance and uncertainty vary greatly across AOIs, tasks and FMs. We believe this is a key issue in practice, because there are many design decisions behind each FM and downstream task (input modalities, sampling, architectures, pretraining, etc.) and usually a downstream task designer is aware of and can decide upon a few of them. Through this work, we advocate for the usage of the methodology herein described (large ablations on reference global labels and simple probes), both when publishing new FMs, and to make informed decisions when designing downstream tasks to use them.