park-2024-lda-estimator-preferred
IN premise — summaries/2026/08/24/park-2024-categorical-hierarchical-concepts-sR-references.md
Created 2026-08-25T02:58:26+00:00
The LDA-based direction estimator (eq. 5.2) is preferred over the simple mean vector E(g_w) for estimating ℓ̄_w due to lower variance.
Summary
When the system needs to estimate the best direction for a given condition, using the LDA-based formula gives a more stable answer than simply averaging the raw values, because it produces less scatter from trial to trial. This matters because a noisier direction estimate would make downstream decisions (like where to point, steer, or allocate) less reliable, so picking the lower-variance option keeps the pipeline more predictable.