Federated Learning for Augmented Industrial Retrieval Yields Innovative Insights into Information and Data Privacy

Researchers from the University of Naples Federico II have made a groundbreaking discovery in the field of information technology and data privacy. According to a recent study, deep learning has significantly advanced Industry 4.0 by leveraging data from the Industrial Internet of Things (IIoT). However, traditional frameworks face challenges in handling multimodal industrial data, including scalability, data privacy, and integration efficiency. The research team has proposed a novel approach, Federated Learning for Augmented Industrial Retrieval (FLAIR), which addresses these challenges through federated learning.

Key Takeaways:

• The researchers have introduced an efficient product retrieval framework for e-commerce systems, addressing privacy and performance challenges through federated learning (FL).
• FLAIR is a novel part retrieval system where distributed warehouses collaboratively train a multimodal foundation model, contrastive language-image pretraining (CLIP), by fine-tuning only the Adapter module via FL, ensuring data privacy and efficiency.
• Effective data augmentation strategies are incorporated to enhance the diversity and quality of the training dataset, addressing the limited availability of multimodal industrial data.
• Comprehensive experiments on the industrial language-image dataset (ILID) highlight that FLAIR holds effective privacy safeguards and strong retrieval capabilities.
• An advanced e-commerce recommendation system built on FLAIR showcases its practical effectiveness.
• FLAIR represents the first application of FL for industrial product retrieval, optimizing part searches, inventory management, and customer experience while maintaining data security.

Statistics:

• The research has been peer-reviewed and published in the Ieee Internet of Things Journal.
• The complete code is available at https://github.com/MODAL-UNINA/FLAIR.
• The study has been funded by the Infrastructure for Big Data and Scientific Computing Project (IBiSco) through Call 424-2018-Action II, HPC Cluster, PNRR Project FAIR-Future AI Research, Spoke 3, and the NRRP MUR Program -NextGenerationEU.
• The research team includes Francesco Piccialli, Diletta Chiaro, Pian Qi, and Valeria Mele from the University of Naples Federico II.

Sources:

• Flair: Federated Learning for Augmented Industrial Retrieval. Ieee Internet of Things Journal, 2025;12(19):39338-39345.
• Ieee-inst Electrical Electronics Engineers Inc, 445 Hoes Lane, Piscataway, NJ 08855-4141, USA
• Francesco Piccialli, University of Naples Federico II, Dept Math & Applicat R Caccioppoli, Math Modelling & Data Anal Res Grp Modal, I-80138 Naples, Italy.