Get our free extension to see links to code for papers anywhere online!

Add to Chrome

Add to Firefox

Picture for Marius Kloft

Marius Kloft

Technical University of Kaiserslautern

Unsupervised Anomaly Detection for Auditing Data and Impact of Categorical Encodings

Oct 26, 2022
Ajay Chawda, Stefanie Grimm, Marius Kloft

Figure 1 for Unsupervised Anomaly Detection for Auditing Data and Impact of Categorical Encodings

Figure 2 for Unsupervised Anomaly Detection for Auditing Data and Impact of Categorical Encodings

Figure 3 for Unsupervised Anomaly Detection for Auditing Data and Impact of Categorical Encodings

Figure 4 for Unsupervised Anomaly Detection for Auditing Data and Impact of Categorical Encodings

* This work has been accepted at Proceedings of the Neurips 2022 Workshop on Synthetic Data 4ML

Access Paper or Ask Questions

Training Normalizing Flows from Dependent Data

Sep 29, 2022
Matthias Kirchler, Christoph Lippert, Marius Kloft

Figure 1 for Training Normalizing Flows from Dependent Data

Figure 2 for Training Normalizing Flows from Dependent Data

Figure 3 for Training Normalizing Flows from Dependent Data

Figure 4 for Training Normalizing Flows from Dependent Data

Access Paper or Ask Questions

Raising the Bar in Graph-level Anomaly Detection

May 27, 2022
Chen Qiu, Marius Kloft, Stephan Mandt, Maja Rudolph

Figure 1 for Raising the Bar in Graph-level Anomaly Detection

Figure 2 for Raising the Bar in Graph-level Anomaly Detection

Figure 3 for Raising the Bar in Graph-level Anomaly Detection

Figure 4 for Raising the Bar in Graph-level Anomaly Detection

* To appear in IJCAI-ECAI 2022

Access Paper or Ask Questions

Exposing Outlier Exposure: What Can Be Learned From Few, One, and Zero Outlier Images

May 23, 2022
Philipp Liznerski, Lukas Ruff, Robert A. Vandermeulen, Billy Joe Franks, Klaus-Robert Müller, Marius Kloft

Figure 1 for Exposing Outlier Exposure: What Can Be Learned From Few, One, and Zero Outlier Images

Figure 2 for Exposing Outlier Exposure: What Can Be Learned From Few, One, and Zero Outlier Images

Figure 3 for Exposing Outlier Exposure: What Can Be Learned From Few, One, and Zero Outlier Images

Figure 4 for Exposing Outlier Exposure: What Can Be Learned From Few, One, and Zero Outlier Images

* 37 pages, preprint. arXiv admin note: text overlap with arXiv:2006.00339

Access Paper or Ask Questions

Detecting Anomalies within Time Series using Local Neural Transformations

Feb 08, 2022
Tim Schneider, Chen Qiu, Marius Kloft, Decky Aspandi Latif, Steffen Staab, Stephan Mandt, Maja Rudolph

Access Paper or Ask Questions