Overcoming Long Inference Time of Nearest Neighbors Analysis in Regression and Uncertainty Prediction

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F68407700%3A21240%2F24%3A00375001" target="_blank" >RIV/68407700:21240/24:00375001 - isvavai.cz</a>

Výsledek na webu

<a href="https://doi.org/10.1007/s42979-024-02670-2" target="_blank" >https://doi.org/10.1007/s42979-024-02670-2</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.1007/s42979-024-02670-2" target="_blank" >10.1007/s42979-024-02670-2</a>

Jazyk výsledku

angličtina

Název v původním jazyce

Overcoming Long Inference Time of Nearest Neighbors Analysis in Regression and Uncertainty Prediction

Popis výsledku v původním jazyce

The intuitive approach of comparing like with like, forms the basis of the so-called nearest neighbor analysis, which is central to many machine learning algorithms. Nearest neighbor analysis is easy to interpret, analyze, and reason about. It is widely used in advanced techniques such as uncertainty estimation in regression models, as well as the renowned k-nearest neighbor-based algorithms. Nevertheless, its high inference time complexity, which is dataset size dependent even in the case of its faster approximated version, restricts its applications and can considerably inflate the application cost. In this paper, we address the problem of high inference time complexity. By using gradient-boosted regression trees as a predictor of the labels obtained from nearest neighbor analysis, we demonstrate a significant increase in inference speed, improving by several orders of magnitude. We validate the effectiveness of our approach on a real-world European Car Pricing Dataset with approximately rows for both residual cost and price uncertainty prediction. Moreover, we assess our method’s performance on the most commonly used tabular benchmark datasets to demonstrate its scalability. The link is to github repository where the code is available: https://github.com/koutefra/uncertainty_experiments.

Název v anglickém jazyce

Overcoming Long Inference Time of Nearest Neighbors Analysis in Regression and Uncertainty Prediction

Popis výsledku anglicky

The intuitive approach of comparing like with like, forms the basis of the so-called nearest neighbor analysis, which is central to many machine learning algorithms. Nearest neighbor analysis is easy to interpret, analyze, and reason about. It is widely used in advanced techniques such as uncertainty estimation in regression models, as well as the renowned k-nearest neighbor-based algorithms. Nevertheless, its high inference time complexity, which is dataset size dependent even in the case of its faster approximated version, restricts its applications and can considerably inflate the application cost. In this paper, we address the problem of high inference time complexity. By using gradient-boosted regression trees as a predictor of the labels obtained from nearest neighbor analysis, we demonstrate a significant increase in inference speed, improving by several orders of magnitude. We validate the effectiveness of our approach on a real-world European Car Pricing Dataset with approximately rows for both residual cost and price uncertainty prediction. Moreover, we assess our method’s performance on the most commonly used tabular benchmark datasets to demonstrate its scalability. The link is to github repository where the code is available: https://github.com/koutefra/uncertainty_experiments.

Druh

J_SC - Článek v periodiku v databázi SCOPUS

CEP obor

—

OECD FORD obor

10201 - Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)

Projekt

—

Návaznosti

S - Specificky vyzkum na vysokych skolach<br>I - Institucionalni podpora na dlouhodoby koncepcni rozvoj vyzkumne organizace

Rok uplatnění

2024

Kód důvěrnosti údajů

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Název periodika

SN Computer Science

ISSN

2662-995X

e-ISSN

2661-8907

Svazek periodika

5

Číslo periodika v rámci svazku

5

Stát vydavatele periodika

SG - Singapurská republika

Počet stran výsledku

12

Strana od-do

—

Kód UT WoS článku

—

EID výsledku v databázi Scopus

2-s2.0-85191305190