-
Federated Learning for Autoencoder-based Condition Monitoring in the Industrial Internet of Things
Authors:
Soeren Becker,
Kevin Styp-Rekowski,
Oliver Vincent Leon Stoll,
Odej Kao
Abstract:
Enabled by the increasing availability of sensor data monitored from production machinery, condition monitoring and predictive maintenance methods are key pillars for an efficient and robust manufacturing production cycle in the Industrial Internet of Things. The employment of machine learning models to detect and predict deteriorating behavior by analyzing a variety of data collected across sever…
▽ More
Enabled by the increasing availability of sensor data monitored from production machinery, condition monitoring and predictive maintenance methods are key pillars for an efficient and robust manufacturing production cycle in the Industrial Internet of Things. The employment of machine learning models to detect and predict deteriorating behavior by analyzing a variety of data collected across several industrial environments shows promising results in recent works, yet also often requires transferring the sensor data to centralized servers located in the cloud. Moreover, although collaborating and sharing knowledge between industry sites yields large benefits, especially in the area of condition monitoring, it is often prohibited due to data privacy issues. To tackle this situation, we propose an Autoencoder-based Federated Learning method utilizing vibration sensor data from rotating machines, that allows for a distributed training on edge devices, located on-premise and close to the monitored machines. Preserving data privacy and at the same time exonerating possibly unreliable network connections of remote sites, our approach enables knowledge transfer across organizational boundaries, without sharing the monitored data. We conducted an evaluation utilizing two real-world datasets as well as multiple testbeds and the results indicate that our method enables a competitive performance compared to previous results, while significantly reducing the resource and network utilization.
△ Less
Submitted 14 November, 2022;
originally announced November 2022.
-
Nesterov Accelerated ADMM for Fast Diffeomorphic Image Registration
Authors:
Alexander Thorley,
Xi Jia,
Hyung ** Chang,
Boyang Liu,
Karina Bunting,
Victoria Stoll,
Antonio de Marvao,
Declan P. O'Regan,
Georgios Gkoutos,
Dipak Kotecha,
**ming Duan
Abstract:
Deterministic approaches using iterative optimisation have been historically successful in diffeomorphic image registration (DiffIR). Although these approaches are highly accurate, they typically carry a significant computational burden. Recent developments in stochastic approaches based on deep learning have achieved sub-second runtimes for DiffIR with competitive registration accuracy, offering…
▽ More
Deterministic approaches using iterative optimisation have been historically successful in diffeomorphic image registration (DiffIR). Although these approaches are highly accurate, they typically carry a significant computational burden. Recent developments in stochastic approaches based on deep learning have achieved sub-second runtimes for DiffIR with competitive registration accuracy, offering a fast alternative to conventional iterative methods. In this paper, we attempt to reduce this difference in speed whilst retaining the performance advantage of iterative approaches in DiffIR. We first propose a simple iterative scheme that functionally composes intermediate non-stationary velocity fields to handle large deformations in images whilst guaranteeing diffeomorphisms in the resultant deformation. We then propose a convex optimisation model that uses a regularisation term of arbitrary order to impose smoothness on these velocity fields and solve this model with a fast algorithm that combines Nesterov gradient descent and the alternating direction method of multipliers (ADMM). Finally, we leverage the computational power of GPU to implement this accelerated ADMM solver on a 3D cardiac MRI dataset, further reducing runtime to less than 2 seconds. In addition to producing strictly diffeomorphic deformations, our methods outperform both state-of-the-art deep learning-based and iterative DiffIR approaches in terms of dice and Hausdorff scores, with speed approaching the inference time of deep learning-based methods.
△ Less
Submitted 26 September, 2021;
originally announced September 2021.
-
Digitizing analogic spectrograms recorded by the Nançay Decameter Array on 35 mm film rolls from 1970 to 1990
Authors:
Baptiste Cecconi,
Laurent Lamy,
Laurent Denis,
Philippe Zarka,
Agnès Fave,
Marie-Pierre Issartel,
Marie-Agnès Dubos,
Corentin Louis,
Pierre Le Sidaner,
Véronique Stoll
Abstract:
The Nançay Decameter Array (NDA), which has now passed 40 years old, acquires daily observations of Jovian and Solar low frequency radio emissions over a continuous spectrum ranging from 10 up to 100MHz, forming the largest database of LW radio observations of these two bodies. It also intermittently observed intense radio sources since its opening in 1977. Before that date, decametric observation…
▽ More
The Nançay Decameter Array (NDA), which has now passed 40 years old, acquires daily observations of Jovian and Solar low frequency radio emissions over a continuous spectrum ranging from 10 up to 100MHz, forming the largest database of LW radio observations of these two bodies. It also intermittently observed intense radio sources since its opening in 1977. Before that date, decametric observations were conducted on the same site with an interferometer formed of a pair of log-periodic Yagi antennas mounted on mobile booms. These observations have been recorded with a series of analogic recorders (before 1990) and then digital receivers (after 1990), with increasing performances and sensitivities. The NDA scientific team recently retrieved and inventoried the archives of analogic data (35mm film rolls) covering two decades (1970 to 1990). We now plan to digitize those observations, in order to recover their scientific value and to include them into the currently operational database covering a time span starting in 1990 up to now, still adding new files every day. This modern and interoperable database has virtual observatory interfaces. It is a required element to foster scientific data exploitation, including Jovian and Solar data analysis over long timescales. We present the status of this project.
△ Less
Submitted 27 March, 2020;
originally announced March 2020.
