Skip to content

Physically-Grounded Variable Selection (PGVS)

Home / Concepts / Physically-Grounded Variable Selection (PGVS)

Physically-Grounded Variable Selection (PGVS)

Auto-generated stub. Edit this file to add more details.

A hierarchical, group-aware sparse selection module that utilizes physical priors to extract condition-adaptive exogenous contexts.

Why It Matters

Provides a novel, physically-informed approach to feature selection in time-series forecasting, addressing the limitations of generic channel mixing or PCA.

Evidence

Physically-Grounded Variable Selection (PGVS) performs hierarchical, group-aware sparse selection over exogenous variables using a domain-informed physical prior and sparsemax activations

Metadata & Links