Summary of Distribution Shifts at Scale: Out-of-distribution Detection in Earth Observation, by Burak Ekim et al.
Distribution Shifts at Scale: Out-of-distribution Detection in Earth Observation
by Burak Ekim, Girmaw Abebe Tadesse, Caleb Robinson, Gilles Hacheme, Michael Schmitt, Rahul Dodhia, Juan M. Lavista Ferres
First submitted to arxiv on: 18 Dec 2024
Categories
- Main: Computer Vision and Pattern Recognition (cs.CV)
- Secondary: Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary In this paper, researchers propose a novel approach to out-of-distribution (OOD) detection for deep learning models in Earth Observation. The method, called TARDIS, is designed for scalable geospatial deployments and addresses the challenge of distribution shifts that degrade model performance. Specifically, TARDIS generates surrogate labels by integrating information from in-distribution data and unknown distributions, allowing OOD detection at scale. The authors validate their approach on two benchmark datasets, EuroSAT and xBD, demonstrating improved performance compared to existing methods. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary The researchers have developed a new way to detect when deep learning models are seeing things they haven’t seen before. This is important for Earth Observation, where images from different places can be very different. The new method, called TARDIS, works by creating fake labels that help the model understand what it’s looking at. The authors tested TARDIS on some big datasets and showed that it performs well. |
Keywords
» Artificial intelligence » Deep learning
