Federated Learning: Challenges, Methods, and Future Directions

Singh, Pushpa; Singh, Murari Kumar; Singh, Rajnesh; Singh, Narendra

doi:10.1007/978-3-030-85559-8_13

Cited by 25 publications

(6 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…where ηkj()t denotes the learning rate, and the first local update is set as θk1()t=θ()t. Depending on the strategy to update the parameters of the global and local models, various techniques have been proposed 44–46 to optimize the communication efficiency compared with the naive SGD method. We used federated averaging ( FedAvg ) in our framework.…”

Section: Methodsmentioning

confidence: 99%

Decentralized Distributed Multi-institutional PET Image Segmentation Using a Federated Deep Learning Framework

Shiri

Sadr²,

Amini

et al. 2022

Clin Nucl Med

View full text Add to dashboard Cite

PurposeThe generalizability and trustworthiness of deep learning (DL)–based algorithms depend on the size and heterogeneity of training datasets. However, because of patient privacy concerns and ethical and legal issues, sharing medical images between different centers is restricted. Our objective is to build a federated DL-based framework for PET image segmentation utilizing a multicentric dataset and to compare its performance with the centralized DL approach.MethodsPET images from 405 head and neck cancer patients from 9 different centers formed the basis of this study. All tumors were segmented manually. PET images converted to SUV maps were resampled to isotropic voxels (3 × 3 × 3 mm3) and then normalized. PET image subvolumes (12 × 12 × 12 cm3) consisting of whole tumors and background were analyzed. Data from each center were divided into train/validation (80% of patients) and test sets (20% of patients). The modified R2U-Net was used as core DL model. A parallel federated DL model was developed and compared with the centralized approach where the data sets are pooled to one server. Segmentation metrics, including Dice similarity and Jaccard coefficients, percent relative errors (RE%) of SUVpeak, SUVmean, SUVmedian, SUVmax, metabolic tumor volume, and total lesion glycolysis were computed and compared with manual delineations.ResultsThe performance of the centralized versus federated DL methods was nearly identical for segmentation metrics: Dice (0.84 ± 0.06 vs 0.84 ± 0.05) and Jaccard (0.73 ± 0.08 vs 0.73 ± 0.07). For quantitative PET parameters, we obtained comparable RE% for SUVmean (6.43% ± 4.72% vs 6.61% ± 5.42%), metabolic tumor volume (12.2% ± 16.2% vs 12.1% ± 15.89%), and total lesion glycolysis (6.93% ± 9.6% vs 7.07% ± 9.85%) and negligible RE% for SUVmax and SUVpeak. No significant differences in performance (P > 0.05) between the 2 frameworks (centralized vs federated) were observed.ConclusionThe developed federated DL model achieved comparable quantitative performance with respect to the centralized DL model. Federated DL models could provide robust and generalizable segmentation, while addressing patient privacy and legal and ethical issues in clinical data sharing.

show abstract

Section: Methodsmentioning

confidence: 99%

Decentralized Distributed Multi-institutional PET Image Segmentation Using a Federated Deep Learning Framework

Shiri

Sadr²,

Amini

et al. 2022

Clin Nucl Med

View full text Add to dashboard Cite

show abstract

“…Instead of directly sharing their raw data, these parties opt for a collaborative process that often involves aggregating or averaging local updates to refine the shared model. For instance, this could apply to situations where different mobile devices collect user behavior data independently, and the device owners wish to maintain data privacy while deriving collective insights through ML [8].…”

Section: Horizontal Flmentioning

confidence: 99%

Federated Learning-based neural networks for autonomous driving

2024

ACE

View full text Add to dashboard Cite

An emerging technology with the capacity to revolutionize the transportation sector is autonomous driving, offering the promise of heightened safety, efficiency, and convenience. However, the widescale deployment of autonomous vehicles presents a multitude of challenges, notably the necessity for robust and adaptable machine learning (ML) models capable of handling a wide array of dynamic real-world scenarios. Enter Federated Learning (FL), a decentralized ML approach that has gained recognition as a potential solution to these challenges. This paper delves into the primary advantages of FL within the context of autonomous driving. It highlights FL's capacity to seamlessly adapt to edge devices, respond to localized changes, and continually enhance safety and performance. The document substantiates these advantages through numerous case studies and empirical evidence, demonstrating how FL can potentially elevate the vision, decision-making, control systems, data transmission, and learning model capabilities of autonomous vehicles. By harnessing the collective intelligence of autonomous vehicles while preserving data privacy and security, FL holds the potential to propel us closer to a future where safe, efficient, and autonomous transportation becomes an attainable reality.

show abstract

“…An insurance company must consider multi-party data to improve its risk management capabilities and commercial expansion. In the insurance industry, effective data utilization without violating individual privacy is a major challenge [46,47].…”

Section: Federated Learning Applicationsmentioning

confidence: 99%

A survey about deep learning and federated Learning in cyberse-curity

Tareq,

Elbagoury,

El-Regaily

et al. 2024

PEN

View full text Add to dashboard Cite

Advances in Artificial Intelligence (AI) technology have led to the strengthening of traditional systems' cybersecurity capabilities in a variety of applications. However, these embedded machine learning models have exposed these systems to a new set of vulnerabilities known as AI assaults. These systems are now attractive targets for cyberattacks, jeopardizing the security and safety of bigger systems that include them. As a result, DL approaches are critical to transitioning network and system protection from providing safe communication between systems to intelligence systems in security. Federated learning (FL) is a new kind of AI based on heterogeneous datasets and decentralized training. FL is a unique research topic that is currently in its early phases. It has not yet gained wide acceptance in the community, owing mostly to privacy and security considerations. In this research, we first shed light on its privacy and security risks that must be discovered, analyzed, and recorded. FL is favored in scenarios where privacy and security are paramount issues. An extensive understanding of risk factors allows an FL adopter and implementer to construct a safe environment successfully while giving researchers a clear perspective of possible study domains. The survey in this paper intends to include an analysis of cybersecurity and DL approaches and modern advances to improve enhanced protection methods. It proposes a complete examination of FL's security and privacy issues to assist in bridging the gap between the current level of federated AI and a future in which broad adoption is achievable. We also propose a range of cybersecurity datasets and the most recently used rating standards.

show abstract

Federated Learning: Challenges, Methods, and Future Directions

Cited by 25 publications

References 14 publications

Decentralized Distributed Multi-institutional PET Image Segmentation Using a Federated Deep Learning Framework

Decentralized Distributed Multi-institutional PET Image Segmentation Using a Federated Deep Learning Framework

Federated Learning-based neural networks for autonomous driving

A survey about deep learning and federated Learning in cyberse-curity

Contact Info

Product

Resources

About