Multiresolution Tensor Learning for Efficient and Interpretable Spatial Analysis
arXiv:2002.05578
Abstract
Efficient and interpretable spatial analysis is crucial in many fields such as geology, sports, and climate science. Tensor latent factor models can describe higher-order correlations for spatial data. However, they are computationally expensive to train and are sensitive to initialization, leading to spatially incoherent, uninterpretable results. We develop a novel Multiresolution Tensor Learning (MRTL) algorithm for efficiently learning interpretable spatial patterns. MRTL initializes the latent factors from an approximate full-rank tensor model for improved interpretability and progressively learns from a coarse resolution to the fine resolution to reduce computation. We also prove the theoretical convergence and computational complexity of MRTL. When applied to two real-world datasets, MRTL demonstrates 4~5x speedup compared to a fixed resolution approach while yielding accurate and interpretable latent factors.
arXiv commentICML 2020
References in corpus (14)
- Towards A Rigorous Science of Interpretable Machine Learning
- Tensor Decomposition for Signal Processing and Machine Learning
- Gaussian Process Optimization in the Bandit Setting: No Regret and Experimental Design
- Tensor Regression with Applications in Neuroimaging Data Analysis
- Identifiability of parameters in latent structure models with many observed variables
- "Why Should I Trust You?": Explaining the Predictions of Any Classifier
- Spatial and Spatio-Temporal Log-Gaussian Cox Processes: Extending the Geostatistical Paradigm
- TensorLy: Tensor Learning in Python
- Parallel Tensor Compression for Large-Scale Scientific Data
- Generating Long-term Trajectories Using Deep Hierarchical Networks
- Multiresolution Recurrent Neural Networks: An Application to Dialogue Response Generation
- Fast and Guaranteed Tensor Decomposition via Sketching
- Tensor Regression Meets Gaussian Processes
- Multiresolution Tensor Decomposition for Multiple Spatial Passing Networks