Fuzzy C-means and explainable AI for quantum entanglement classification and noise analysis

Quantum entanglement plays a fundamental role in quantum mechanics, with applications in quantum computing. This study introduces a new approach that integrates quantum simulations, noise analysis, and fuzzy clustering to classify and evaluate the stability of quantum entangled states under noisy co...

Descripción completa

Detalles Bibliográficos
Autor: Marín Díaz, Gabriel
Tipo de recurso: artículo
Fecha de publicación:2025
País:España
Institución:Universidad Complutense de Madrid (UCM)
Repositorio:Docta Complutense
Idioma:inglés
OAI Identifier:oai:docta.ucm.es:20.500.14352/129673
Acceso en línea:https://hdl.handle.net/20.500.14352/129673
Access Level:acceso abierto
Palabra clave:004.8
311
530.145.8
quantum entanglement
quantum decoherence mitigation
fuzzy clustering in quantum systems
Explainable Artificial Intelligence
XAI
Física de materiales
Teoría de los quanta
Inteligencia artificial (Informática)
Estadística aplicada
2212.12 Teoría Cuántica de Campos
2208.07 Física de Partículas
1203.04 Inteligencia Artificial
1209.01 Estadística Analítica
Descripción
Sumario:Quantum entanglement plays a fundamental role in quantum mechanics, with applications in quantum computing. This study introduces a new approach that integrates quantum simulations, noise analysis, and fuzzy clustering to classify and evaluate the stability of quantum entangled states under noisy conditions. The Fuzzy C-Means clustering model (FCM) is applied to identify different categories of quantum states based on fidelity and entropy trends, allowing for a structured assessment of the impact of noise. The presented methodology follows five key phases: a simulation of the Bell state, the introduction of the noise channel (depolarization and phase damping), noise suppression using corrective operators, clustering-based state classification, and interpretability analysis using Explainable Artificial Intelligence (XAI) techniques. The results indicate that while moderate noise levels allow for partial state recovery, strong decoherence, particularly under depolarization, remains a major challenge. Rather than relying solely on noise suppression, a classification-based strategy is proposed to identify states that retain computational feasibility despite the effects of noise. This hybrid approach combining quantum-state classification with AI-based interpretability offers a new framework for assessing the resilience of quantum systems. The results have practical implications in quantum error correction, quantum cryptography, and the optimization of quantum technologies under realistic conditions.