MQFL-FHE: Multimodal Quantum Federated Learning Framework with Fully Homomorphic Encryption

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Dutta, Siddhant, Innan, Nouhaila, Yahia, Sadok Ben, Shafique, Muhammad, Neira, David Esteban Bernal
Formato: Preprint
Publicado: 2024
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866911290952253440
author Dutta, Siddhant
Innan, Nouhaila
Yahia, Sadok Ben
Shafique, Muhammad
Neira, David Esteban Bernal
author_facet Dutta, Siddhant
Innan, Nouhaila
Yahia, Sadok Ben
Shafique, Muhammad
Neira, David Esteban Bernal
contents The integration of fully homomorphic encryption (FHE) in federated learning (FL) has led to significant advances in data privacy. However, during the aggregation phase, it often results in performance degradation of the aggregated model, hindering the development of robust representational generalization. In this work, we propose a novel multimodal quantum federated learning framework that utilizes quantum computing to counteract the performance drop resulting from FHE. For the first time in FL, our framework combines a multimodal quantum mixture of experts (MQMoE) model with FHE, incorporating multimodal datasets for enriched representation and task-specific learning. Our MQMoE framework enhances performance on multimodal datasets and combined genomics and brain MRI scans, especially for underrepresented categories. Our results also demonstrate that the quantum-enhanced approach mitigates the performance degradation associated with FHE and improves classification accuracy across diverse datasets, validating the potential of quantum interventions in enhancing privacy in FL.
format Preprint
id arxiv_https___arxiv_org_abs_2412_01858
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle MQFL-FHE: Multimodal Quantum Federated Learning Framework with Fully Homomorphic Encryption
Dutta, Siddhant
Innan, Nouhaila
Yahia, Sadok Ben
Shafique, Muhammad
Neira, David Esteban Bernal
Quantum Physics
Cryptography and Security
Distributed, Parallel, and Cluster Computing
Emerging Technologies
Machine Learning
The integration of fully homomorphic encryption (FHE) in federated learning (FL) has led to significant advances in data privacy. However, during the aggregation phase, it often results in performance degradation of the aggregated model, hindering the development of robust representational generalization. In this work, we propose a novel multimodal quantum federated learning framework that utilizes quantum computing to counteract the performance drop resulting from FHE. For the first time in FL, our framework combines a multimodal quantum mixture of experts (MQMoE) model with FHE, incorporating multimodal datasets for enriched representation and task-specific learning. Our MQMoE framework enhances performance on multimodal datasets and combined genomics and brain MRI scans, especially for underrepresented categories. Our results also demonstrate that the quantum-enhanced approach mitigates the performance degradation associated with FHE and improves classification accuracy across diverse datasets, validating the potential of quantum interventions in enhancing privacy in FL.
title MQFL-FHE: Multimodal Quantum Federated Learning Framework with Fully Homomorphic Encryption
topic Quantum Physics
Cryptography and Security
Distributed, Parallel, and Cluster Computing
Emerging Technologies
Machine Learning
url https://arxiv.org/abs/2412.01858