Few-Shot High-Dimensional Feature Selection with Lagrange Programming Neural Networks

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F67985807%3A_____%2F25%3A00643476" target="_blank" >RIV/67985807:_____/25:00643476 - isvavai.cz</a>

Výsledek na webu

<a href="https://doi.org/10.1109/IJCNN64981.2025.11227603" target="_blank" >https://doi.org/10.1109/IJCNN64981.2025.11227603</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.1109/IJCNN64981.2025.11227603" target="_blank" >10.1109/IJCNN64981.2025.11227603</a>

Jazyk výsledku

angličtina

Název v původním jazyce

Few-Shot High-Dimensional Feature Selection with Lagrange Programming Neural Networks

Popis výsledku v původním jazyce

Few-shot learning and high-dimensional feature selection represent significant challenges in machine learning. In this work, we introduce LPNN-FS, a novel feature selection (FS) technique based on a Lagrange Programming Neural Network (Pk-LPNN) originally developed in the context of compressive sampling. LPNN-FS is a continuous-time recurrent neural network whose dynamics inherently converges to an optimal sparse solution of the LASSO, with its equilibrium point acting as a feature selector. Evaluated in combination with specific downstream classifiers, our model is tested on both synthetic and real-world benchmark datasets characterized by high dimensionality and a limited number of observations. The results demonstrate that LPNN-FS competes with or outperforms state-of-the-art methods such as LASSO, k-best, and sparse PCA in this few-shot, high-dimensional setting. Overall, this study opens new avenues for leveraging compressive sampling-based techniques in challenging feature selection problems.

Název v anglickém jazyce

Few-Shot High-Dimensional Feature Selection with Lagrange Programming Neural Networks

Popis výsledku anglicky

Few-shot learning and high-dimensional feature selection represent significant challenges in machine learning. In this work, we introduce LPNN-FS, a novel feature selection (FS) technique based on a Lagrange Programming Neural Network (Pk-LPNN) originally developed in the context of compressive sampling. LPNN-FS is a continuous-time recurrent neural network whose dynamics inherently converges to an optimal sparse solution of the LASSO, with its equilibrium point acting as a feature selector. Evaluated in combination with specific downstream classifiers, our model is tested on both synthetic and real-world benchmark datasets characterized by high dimensionality and a limited number of observations. The results demonstrate that LPNN-FS competes with or outperforms state-of-the-art methods such as LASSO, k-best, and sparse PCA in this few-shot, high-dimensional setting. Overall, this study opens new avenues for leveraging compressive sampling-based techniques in challenging feature selection problems.

Druh

D - Stať ve sborníku

CEP obor

—

OECD FORD obor

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

Projekt

<a href="/cs/project/GA25-15490S" target="_blank" >GA25-15490S: LEDNeCo: Nízkoenergetické hluboké neurovýpočty</a><br>

Návaznosti

I - Institucionalni podpora na dlouhodoby koncepcni rozvoj vyzkumne organizace

Rok uplatnění

2025

Kód důvěrnosti údajů

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

Název statě ve sborníku

IJCNN 2025: International Joint Conference on Neural Networks Conference Proceedings

ISBN

979-8-3315-1042-8

ISSN

—

e-ISSN

—

Počet stran výsledku

8

Strana od-do

—

Název nakladatele

IEEE

Místo vydání

Piscataway

Místo konání akce

Rome

Datum konání akce

30. 6. 2025

Typ akce podle státní příslušnosti

WRD - Celosvětová akce

Kód UT WoS článku

—