Streaming Gaussian Dirichlet Random Fields for Spatial Predictions of High Dimensional Categorical Observations
This enables efficient informative path planning for high-dimensional categorical observations, addressing a previously infeasible task in spatial prediction.
The paper tackles the problem of modeling streaming spatiotemporal high-dimensional categorical data by introducing the S-GDRF model, which achieves more accurate predictions than a Variational Gaussian Process on plankton image data.
We present the Streaming Gaussian Dirichlet Random Field (S-GDRF) model, a novel approach for modeling a stream of spatiotemporally distributed, sparse, high-dimensional categorical observations. The proposed approach efficiently learns global and local patterns in spatiotemporal data, allowing for fast inference and querying with a bounded time complexity. Using a high-resolution data series of plankton images classified with a neural network, we demonstrate the ability of the approach to make more accurate predictions compared to a Variational Gaussian Process (VGP), and to learn a predictive distribution of observations from streaming categorical data. S-GDRFs open the door to enabling efficient informative path planning over high-dimensional categorical observations, which until now has not been feasible.