Anomaly Detection Using Federated Learning

Singh, Shubham; Bhardwaj, Shantanu; Pandey, Hemlatha; Beniwal, Gunjan

doi:10.1007/978-981-15-4992-2_14

Cited by 8 publications

(2 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…More recently, federated anomaly detection draws increasing attention [44][45][46][47][48]. Nguyen et al propose to apply federated anomaly detection for IoT devices [45].…”

Section: Federated Learningmentioning

confidence: 99%

Deep Federated Anomaly Detection for Multivariate Time Series Data

Zhu¹,

Song²,

Chen³

et al. 2022

Preprint

View full text Add to dashboard Cite

Despite the fact that many anomaly detection approaches have been developed for multivariate time series data, limited effort has been made on federated settings in which multivariate time series data are heterogeneously distributed among different edge devices while data sharing is prohibited. In this paper, we investigate the problem of federated unsupervised anomaly detection and present a Federated Exemplarbased Deep Neural Network (Fed-ExDNN) to conduct anomaly detection for multivariate time series data on different edge devices. Specifically, we first design an Exemplar-based Deep Neural network (ExDNN) to learn local time series representations based on their compatibility with an exemplar module which consists of hidden parameters learned to capture varieties of normal patterns on each edge device. Next, a constrained clustering mechanism (FedCC) is employed on the centralized server to align and aggregate the parameters of different local exemplar modules to obtain a unified global exemplar module. Finally, the global exemplar module is deployed together with a shared feature encoder to each edge device and anomaly detection is conducted by examining the compatibility of testing data to the exemplar module. Fed-ExDNN captures local normal time series patterns with ExDNN and aggregates these patterns by FedCC, and thus can handle the heterogeneous data distributed over different edge devices simultaneously. Thoroughly empirical studies on six public datasets show that ExDNN and Fed-ExDNN can outperform state-ofthe-art anomaly detection algorithms and federated learning techniques.

show abstract

“…More recently, federated anomaly detection draws increasing attention [44][45][46][47][48]. Nguyen et al propose to apply federated anomaly detection for IoT devices [45].…”

Section: Federated Learningmentioning

confidence: 99%

Deep Federated Anomaly Detection for Multivariate Time Series Data

Zhu¹,

Song²,

Chen³

et al. 2022

Preprint

View full text Add to dashboard Cite

show abstract

“…All organizations are using or planning to use AI in different ways. The pharmaceutical industry has been benefitted a lot as it helps in early detection, clinical developments, ease of diagnosis, and many more cases (Singh et al, 2021). AI has also helped in managing the impact of COVID-19 through tracking and prediction, treatments, vaccines, diagnosis and social control (Naud e, 2020).…”

Section: Augmented Reality (Ar)mentioning

confidence: 99%

Procurement 4.0 to the rescue: catalysing its adoption by modelling the challenges

Jerome

Saxena

Sonwaney

et al. 2021

BIJ

View full text Add to dashboard Cite

PurposeThe pandemic crisis has resulted in global chaos that had caused massive disruption to the supply chain. The pharmaceutical industry, in particular, has been working tirelessly to ensure that they can cater to the people who need them. With restrictions being imposed to prevent the spread of the COVID-19 virus, the movement of raw materials required has been affected, thus creating the need for the procurement function to be innovative. This study proposes the application of Industry 4.0 concepts into the procurement activities of an organization to make it more resilient and efficient.Design/methodology/approachTo study the intensity of the challenges, Total Interpretive Structural Modelling is used alongside the “Matrice des Impacts Croises Multiplication Appliquee a un Classement” (MICMAC) technique.FindingsResilience can be achieved through the collaboration between the organization and its network of suppliers. This is however easier said than done. High and unclear investments have been identified as the challenge that is taking a toll on all technological investments in the pandemic era. The study also shows that organizational inertia which is present in established and structured firms are a deterrent as well.Originality/valueThis study is based on the application of procurement 4.0 to ensure that pharmaceutical supply chains stay least affected since they are essentials. This study using a multi-criteria decision-making approach to prioritize the challenges. This will help practitioners make decisions faster.

show abstract