TIML: Task-Informed Meta-Learning for Agriculture

TL;DR

This paper introduces TIML, a meta-learning approach that incorporates task-specific metadata to improve agricultural data analysis in data-sparse regions, demonstrating significant performance gains over benchmarks.

Contribution

TIML extends model-agnostic meta-learning by integrating task-specific metadata, enhancing transfer learning effectiveness in agriculture and potentially other geospatial applications.

Findings

TIML outperforms benchmarks in crop classification.

TIML improves yield estimation accuracy.

Applicable to diverse models and tasks.

Abstract

Labeled datasets for agriculture are extremely spatially imbalanced. When developing algorithms for data-sparse regions, a natural approach is to use transfer learning from data-rich regions. While standard transfer learning approaches typically leverage only direct inputs and outputs, geospatial imagery and agricultural data are rich in metadata that can inform transfer learning algorithms, such as the spatial coordinates of data-points or the class of task being learned. We build on previous work exploring the use of meta-learning for agricultural contexts in data-sparse regions and introduce task-informed meta-learning (TIML), an augmentation to model-agnostic meta-learning which takes advantage of task-specific metadata. We apply TIML to crop type classification and yield estimation, and find that TIML significantly improves performance compared to a range of benchmarks in both contexts, across a diversity of model architectures. While we focus on tasks from agriculture, TIML could offer benefits to any meta-learning setup with task-specific metadata, such as classification of geo-tagged images and species distribution modelling.