Intrinsic Gaussian Process Regression Modeling for Manifold-Valued Response Variables

A 2026 methodological paper by Xinyu Li proposes a novel intrinsic Gaussian process regression model for data with a response variable measured on a Riemannian manifold. The work includes numerical studies with simulations and real examples to validate the method. The dataset, shared on figshare under a CC-BY-4.0 license, includes supporting files in PDF, ZIP, and TXT formats totaling 2.0 MB.

Use Cases

• Developing intrinsic Gaussian process regression models based on the described Riemannian manifold framework.
• Comparing intrinsic and extrinsic methods for manifold-valued data analysis as discussed in the paper.
• Applying parallel transport operators to define covariance structures for non-Euclidean data.
• Studying asymptotic properties of posterior distribution estimation for regression on manifolds.
• Implementing kernel methods for data situated on manifolds without a natural ambient space.

Strengths

• The method is validated with numerical studies including simulation and real examples.
• The model is designed to be invariant to the choice of orthonormal frames for coordinate representations.
• The dataset is openly shared under a permissive CC-BY-4.0 license.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.
• The 2.0 MB size suggests a small dataset, potentially limiting the scope of included examples.

Provenance

Source

figshare

Collection Method

Likely contains data generated for numerical studies and real examples supporting the academic paper.

Freshness

Last updated 2026-05-12 21:41:30; freshness should be verified.