Red Teaming Models for Hyperspectral Image Analysis Using Explainable AI (2403.08017v2)
Abstract: Remote sensing (RS) applications in the space domain demand ML models that are reliable, robust, and quality-assured, making red teaming a vital approach for identifying and exposing potential flaws and biases. Since both fields advance independently, there is a notable gap in integrating red teaming strategies into RS. This paper introduces a methodology for examining ML models operating on hyperspectral images within the HYPERVIEW challenge, focusing on soil parameters' estimation. We use post-hoc explanation methods from the Explainable AI (XAI) domain to critically assess the best performing model that won the HYPERVIEW challenge and served as an inspiration for the model deployed on board the INTUITION-1 hyperspectral mission. Our approach effectively red teams the model by pinpointing and validating key shortcomings, constructing a model that achieves comparable performance using just 1% of the input features and a mere up to 5% performance loss. Additionally, we propose a novel way of visualizing explanations that integrate domain-specific information about hyperspectral bands (wavelengths) and data transformations to better suit interpreting models for hyperspectral image analysis.
- Towards explainable ai4eo: An explainable deep learning approach for crop type mapping using satellite images time series. In IGARSS 2023-2023 IEEE International Geoscience and Remote Sensing Symposium, pp. 1088–1091. IEEE, 2023.
- Explanatory Model Analysis. Chapman and Hall/CRC, New York, 2021.
- A survey on deep neural network pruning-taxonomy, comparison, analysis, and recommendations. arXiv preprint arXiv:2308.06767, 2023.
- Understanding Global Feature Contributions With Additive Importance Measures. In NeurIPS, 2020.
- Counterfactual Explanations for Remote Sensing Time Series Data: An Application to Land Cover Classification. In ECML PKDD, 2023.
- Towards explainable ai for hyperspectral image classification in edge computing environments. Computers and Electrical Engineering, 103:108381, 2022.
- Explaining Multimodal Data Fusion: Occlusion Analysis for Wilderness Mapping. In ICLR ML4RS Workshop, 2023.
- Confident naturalness explanation (cne): A framework to explain and assess patterns forming naturalness in fennoscandia with confidence. In Northern Lights Deep Learning Conference, number FZJ-2024-00529. Pflanzenwissenschaften, 2023a.
- Leveraging activation maximization and generative adversarial training to recognize and explain patterns in natural areas in satellite imagery. IEEE Geoscience and Remote Sensing Letters, 2023b.
- Red teaming language models to reduce harms: Methods, scaling behaviors, and lessons learned. arXiv preprint arXiv:2209.07858, 2022.
- Caroline M Gevaert. Explainable ai for earth observation: A review including societal and regulatory perspectives. International Journal of Applied Earth Observation and Geoinformation, 112:102869, 2022.
- Network pruning for remote sensing images classification based on interpretable cnns. IEEE Transactions on Geoscience and Remote Sensing, 60:1–15, 2021.
- Explainable AI Methods – A Brief Overview. In xxAI - Beyond Explainable AI: International Workshop, Held in Conjunction with ICML 2020, pp. 13–38, 2022.
- Predicting soil properties from hyperspectral satellite images. In 2022 IEEE International Conference on Image Processing (ICIP), pp. 4296–4300. IEEE, 2022.
- A Unified Approach to Interpreting Model Predictions. In NeurIPS, 2017.
- From local explanations to global understanding with explainable ai for trees. Nature machine intelligence, 2(1):56–67, 2020.
- Flirt: Feedback loop in-context red teaming. arXiv preprint arXiv:2308.04265, 2023.
- The hyperview challenge: Estimating soil parameters from hyperspectral images. In 2022 IEEE International Conference on Image Processing (ICIP), pp. 4268–4272. IEEE, 2022.
- Red Teaming Language Models with Language Models. In EMNLP, 2022.
- A network pruning method for remote sensing image scene classification. In 2019 IEEE International Conference on Signal, Information and Data Processing (ICSIDP), pp. 1–4. IEEE, 2019.
- Red-Teaming the Stable Diffusion Safety Filter. In NeurIPS ML Safety Workshop, 2022.
- Explaining classifications for individual instances. IEEE Transactions on Knowledge and Data Engineering, 20(5):589–600, 2008.
- Explain it to me–facing remote sensing challenges in the bio-and geosciences with explainable machine learning. ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences, 3:817–824, 2020a.
- Explainable machine learning for scientific insights and discoveries. Ieee Access, 8:42200–42216, 2020b.
- Estimation of chlorophyll, macronutrients and water content in maize from hyperspectral data using machine learning and explainable artificial intelligence techniques. Remote Sensing Letters, 13(10):969–979, 2022.
- Interpreting hyperspectral remote sensing image classification methods via explainable artificial intelligence. In IGARSS 2023-2023 IEEE International Geoscience and Remote Sensing Symposium, pp. 5950–5953. IEEE, 2023.
- Taking artificial intelligence into space through objective selection of hyperspectral earth observation applications: To bring the “brain” close to the “eyes” of satellite missions. IEEE Geoscience and Remote Sensing Magazine, 11(2):10–39, 2023.
- Trees, forests, chickens, and eggs: when and why to prune trees in a random forest. Statistical Analysis and Data Mining: The ASA Data Science Journal, 16(1):45–64, 2023.
- Feature selection method based on high-resolution remote sensing images and the effect of sensitive features on classification accuracy. Sensors, 18(7):2013, 2018.
Sponsor
Paper Prompts
Sign up for free to create and run prompts on this paper using GPT-5.
Top Community Prompts
Collections
Sign up for free to add this paper to one or more collections.