On-device research index

arXiv research

A locally-built, LLM-digested index of recent arXiv papers in quant finance, geometry/topology, and statistical ML — keyword search served straight from SQLite on this machine.

168,742 papers · 148 categories

Trend · papers per month

count % of corpus

3 results for “average-K”

Classification Under Ambiguity: When Is Average-K Better Than Top-K?stat.ML

This paper compares average-K and top-K classification methods under ambiguity.

problem Choosing a single label in ambiguous cases leads to low precision.

method Formally characterizes ambiguity profiles and compares average-K and top-K classification methods.

result Average-K can achieve lower error rates than top-K in some ambiguous cases.

classification ambiguity average-K top-K error rate

2021-12-16abs ↗pdf ↗

Learning Efficient Anomaly Detectors from

K

-NN Graphscs.LG

We propose a non-parametric anomaly detection algorithm for high dimensional data. We score each datapoint by its average $K$ -NN distance, and rank them accordingly. We then train limited complexity models to imitate these scores based on the max-margin learning-to-rank framework. A test-point is declared as an anomaly…

2015-02-06abs ↗pdf ↗

On index expectation curvature for manifoldsmath.DG

New curvature K(x) measures manifold properties without integrals.

problem Understanding curvature on compact Riemannian manifolds.

method Developed index expectation curvature K(x) for 2D manifolds, constructed as a product of sectional index expectation curvatures.

result For small 2D manifolds with boundary, definite sign index expectation curvature K(x) exists and satisfies Gauss-Bonnet relation.

manifold curvature index expectation Gauss-Bonnet Morse functions sectional curvature

2020-01-20abs ↗pdf ↗