UniLVQ: A Unified Learning Vector Quantization Framework for Supervised Learning Tasks

UniLVQ: A Unified Learning Vector Quantization Framework for Supervised Learning Tasks

<p dir="ltr">UniLVQ is a unified and extensible framework for prototype-based learning using Learning Vector Quantization (LVQ) and its modern variants. Built with PyTorch and wrapped in scikit-learn's BaseEstimator interface, UniLVQ supports both rule-based methods (e.g., LVQ1,...

Full description

Saved in:
Bibliographic Details
Main Author: Van Thieu Nguyen (21096587) (author)
Published: 2025
Subjects:
Evolutionary computation
Modelling and simulation
Artificial intelligence not elsewhere classified
Distributed systems and algorithms
High performance computing
Performance evaluation
Distributed computing and systems software not elsewhere classified
Open access
Neural networks
Machine learning not elsewhere classified
Automated software engineering
Software architecture
Software engineering not elsewhere classified
Prototype learning
learning vector quantization (LVQ)
learning vector quantization networks
Rule-based modelling
gradient descent search
Supervised Learning
Machine Learning
Regression
Classification
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:<p dir="ltr">UniLVQ is a unified and extensible framework for prototype-based learning using Learning Vector Quantization (LVQ) and its modern variants. Built with PyTorch and wrapped in scikit-learn's BaseEstimator interface, UniLVQ supports both rule-based methods (e.g., LVQ1, LVQ2.1, LVQ3) and loss-based gradient-optimized variants (e.g., GLVQ, GRLVQ, RSLVQ). The framework is designed for supervised learning problems, including both classification and regression tasks. UniLVQ enables easy integration with scikit-learn pipelines and offers modularity, interpretability, and support for future extensions such as metaheuristic-based optimization.</p>

Similar Items