Better, Not Just More: Data-Centric Machine Learning for Earth Observation
arXiv:2312.05327 · doi:10.1109/MGRS.2024.3470986
Abstract
Recent developments and research in modern machine learning have led to substantial improvements in the geospatial field. Although numerous deep learning architectures and models have been proposed, the majority of them have been solely developed on benchmark datasets that lack strong real-world relevance. Furthermore, the performance of many methods has already saturated on these datasets. We argue that a shift from a model-centric view to a complementary data-centric perspective is necessary for further improvements in accuracy, generalization ability, and real impact on end-user applications. Furthermore, considering the entire machine learning cycle-from problem definition to model deployment with feedback-is crucial for enhancing machine learning models that can be reliable in unforeseen situations. This work presents a definition as well as a precise categorization and overview of automated data-centric learning approaches for geospatial data. It highlights the complementary role of data-centric learning with respect to model-centric in the larger machine learning deployment cycle. We review papers across the entire geospatial field and categorize them into different groups. A set of representative experiments shows concrete implementation examples. These examples provide concrete steps to act on geospatial data with data-centric machine learning approaches.
References in corpus (26)
- SpectralGPT: Spectral Remote Sensing Foundation Model
- A survey of active learning algorithms for supervised remote sensing image classification
- No Classification without Representation: Assessing Geodiversity Issues in Open Data Sets for the Developing World
- OpenStreetMap: Challenges and Opportunities in Machine Learning and Remote Sensing
- SEN12MS-CR-TS: A Remote Sensing Data Set for Multi-modal Multi-temporal Cloud Removal
- The Principles of Data-Centric AI (DCAI)
- From Easy to Hard: Learning Language-guided Curriculum for Visual Question Answering on Remote Sensing Data
- On the Opportunities and Challenges of Foundation Models for Geospatial Artificial Intelligence
- On the Effects of Different Types of Label Noise in Multi-Label Remote Sensing Image Classification
- There Are No Data Like More Data- Datasets for Deep Learning in Earth Observation
- Domino: Discovering Systematic Errors with Cross-Modal Embeddings
- SSL4EO-S12: A Large-Scale Multi-Modal, Multi-Temporal Dataset for Self-Supervised Learning in Earth Observation
- A Distributional Framework for Data Valuation
- Evaluation Challenges for Geospatial ML
- Mission Critical -- Satellite Data is a Distinct Modality in Machine Learning
- Reliability Scores from Saliency Map Clusters for Improved Image-based Harvest-Readiness Prediction in Cauliflower
- Data Banzhaf: A Robust Data Valuation Framework for Machine Learning
- Categorising the World into Local Climate Zones -- Towards Quantifying Labelling Uncertainty for Machine Learning Models
- Estimating Physical Information Consistency of Channel Data Augmentation for Remote Sensing Images
- Perspectives on the State and Future of Deep Learning - 2023
- HyperSIGMA: Hyperspectral Intelligence Comprehension Foundation Model
- A Comprehensive Study on the Robustness of Image Classification and Object Detection in Remote Sensing: Surveying and Benchmarking
- Responsible AI for Earth Observation
- Efficient Curriculum based Continual Learning with Informative Subset Selection for Remote Sensing Scene Classification
- Impacts of Color and Texture Distortions on Earth Observation Data in Deep Learning
- A survey of Identification and mitigation of Machine Learning algorithmic biases in Image Analysis