Search Results for author: Gabriel Michau

Found 13 papers, 2 papers with code

Learning Informative Health Indicators Through Unsupervised Contrastive Learning

no code implementations28 Aug 2022 Katharina Rombach, Gabriel Michau, Wilfried Bürzle, Stefan Koller, Olga Fink

Our results demonstrate that the proposed approach is able to learn the ground truth health evolution of milling machines and the learned health indicator is suited for fault detection of railway wheels operated under various operating conditions by outperforming state-of-the-art methods.

Anomaly Detection Contrastive Learning +2

Fully Learnable Deep Wavelet Transform for Unsupervised Monitoring of High-Frequency Time Series

no code implementations3 May 2021 Gabriel Michau, Gaetan Frusque, Olga Fink

In this paper, we propose a fully unsupervised deep learning framework that is able to extract a meaningful and sparse representation of raw HF signals.

Denoising Time Series +1

Improving Generalization of Deep Fault Detection Models in the Presence of Mislabeled Data

no code implementations30 Sep 2020 Katharina Rombach, Gabriel Michau, Olga Fink

In the first step, we identify outliers (including the mislabeled samples) based on the update in the hypothesis space.

Data Augmentation Fault Detection

Unsupervised Transfer Learning for Anomaly Detection: Application to Complementary Operating Condition Transfer

no code implementations18 Aug 2020 Gabriel Michau, Olga Fink

A solution to this problem is to perform unsupervised transfer learning (UTL), to transfer complementary data between different units.

Clustering Dimensionality Reduction +3

Interpretable Detection of Partial Discharge in Power Lines with Deep Learning

no code implementations13 Aug 2020 Gabriel Michau, Chi-Ching Hsu, Olga Fink

It provides interpretability of the results for the domain experts with the identification of the pulses that led to the detection of the PDs.

Temporal signals to images: Monitoring the condition of industrial assets with deep learning image processing algorithms

no code implementations14 May 2020 Gabriel Rodriguez Garcia, Gabriel Michau, Mélanie Ducoffe, Jayant Sen Gupta, Olga Fink

Essential characteristics of time series, situated outside the time domain, are often difficult to capture with state-of-the-art anomaly detection methods when no transformations have been applied to the time series.

Time Series Time Series Analysis +1

Anomaly Detection And Classification In Time Series With Kervolutional Neural Networks

no code implementations14 May 2020 Oliver Ammann, Gabriel Michau, Olga Fink

We demonstrate that mixing kervolutional with convolutional layers in the encoder is more sensitive to variations in the input data and is able to detect anomalous time series in a better way.

Anomaly Detection Classification +5

Missing-Class-Robust Domain Adaptation by Unilateral Alignment for Fault Diagnosis

3 code implementations7 Jan 2020 Qin Wang, Gabriel Michau, Olga Fink

We demonstrate in this paper that the performance of domain adversarial methods can be vulnerable to an incomplete target label space during training.

Domain Adaptation

Domain Adaptation for One-Class Classification: Monitoring the Health of Critical Systems Under Limited Information

no code implementations22 Jul 2019 Gabriel Michau, Olga Fink

In the early life of the system, the collected data is probably not representative of future operating conditions, making it challenging to train a robust model.

Domain Adaptation Fault Detection +4

Unsupervised Fault Detection in Varying Operating Conditions

no code implementations15 Jul 2019 Gabriel Michau, Olga Fink

Two approaches rely on the data from the unit to be monitored only: the baseline is trained on the early life of the unit.

Fault Detection Incremental Learning

Domain Adaptive Transfer Learning for Fault Diagnosis

no code implementations15 May 2019 Qin Wang, Gabriel Michau, Olga Fink

Thanks to digitization of industrial assets in fleets, the ambitious goal of transferring fault diagnosis models fromone machine to the other has raised great interest.

Domain Adaptation Transfer Learning

Feature Learning for Fault Detection in High-Dimensional Condition-Monitoring Signals

1 code implementation12 Oct 2018 Gabriel Michau, Yang Hu, Thomas Palmé, Olga Fink

The performance is first evaluated on a synthetic dataset that encompasses typical characteristics of condition monitoring data.

Dimensionality Reduction Fault Detection +4

Cannot find the paper you are looking for? You can Submit a new open access paper.