DeepAI AI Chat
Log In Sign Up

Active Nearest Neighbor Regression Through Delaunay Refinement

by   Alexander Kravberg, et al.

We introduce an algorithm for active function approximation based on nearest neighbor regression. Our Active Nearest Neighbor Regressor (ANNR) relies on the Voronoi-Delaunay framework from computational geometry to subdivide the space into cells with constant estimated function value and select novel query points in a way that takes the geometry of the function graph into account. We consider the recent state-of-the-art active function approximator called DEFER, which is based on incremental rectangular partitioning of the space, as the main baseline. The ANNR addresses a number of limitations that arise from the space subdivision strategy used in DEFER. We provide a computationally efficient implementation of our method, as well as theoretical halting guarantees. Empirical results show that ANNR outperforms the baseline for both closed-form functions and real-world examples, such as gravitational wave parameter inference and exploration of the latent space of a generative model.


page 7

page 9


Learning to Index for Nearest Neighbor Search

In this study, we present a novel ranking model based on learning the ne...

Nearest neighbor representations of Boolean functions

A nearest neighbor representation of a Boolean function is a set of posi...

Bayesian Kernel and Mutual k-Nearest Neighbor Regression

We propose Bayesian extensions of two nonparametric regression methods w...

Social Distancing is Good for Points too!

The nearest-neighbor rule is a well-known classification technique that,...

Nearest Neighbor Classifier with Margin Penalty for Active Learning

As deep learning becomes the mainstream in the field of natural language...

Unsupervised Space Partitioning for Nearest Neighbor Search

Approximate Nearest Neighbor Search (ANNS) in high dimensional spaces is...

Latent Geometry and Memorization in Generative Models

It can be difficult to tell whether a trained generative model has learn...