This shows you the differences between two versions of the page.

Link to this comparison view

Next revision
Previous revision
anomaly_detection [2017/09/21 15:43] external edit
anomaly_detection [2018/12/06 14:53] (current)
Line 71: Line 71:
 relevant anomalies in the data. We envision this method as an important supplement relevant anomalies in the data. We envision this method as an important supplement
 to the forensic examiners’ toolbox. to the forensic examiners’ toolbox.
 +https://​arxiv.org/​abs/​1808.05492v1 Metric Learning for Novelty and Anomaly Detection
 +We show that metric learning
 +provides a better output embedding space to detect data outside the learned distribution
 +than cross-entropy softmax based models. This opens an opportunity to further research on
 +how this embedding space should be learned, with restrictions that could further improve the
 +field. The presented results suggest that out-of-distribution data might not all be seen as a
 +single type of anomaly, but instead a continuous representation between novelty and anomaly
 +data. In that spectrum, anomaly detection is the easier task, giving more focus at the difficulty
 +of novelty detection. ​ https://​mmasana.github.io/​OoD_Mining/​
 +https://​arxiv.org/​abs/​1802.04865 Learning Confidence for Out-of-Distribution Detection in Neural Networks https://​github.com/​ShiyuLiang/​odin-pytorch
 +https://​arxiv.org/​abs/​1809.04758 Anomaly Detection with Generative Adversarial Networks for Multivariate Time Series
 +https://​arxiv.org/​abs/​1810.01392v1 Generative Ensembles for Robust Anomaly Detection
 +we propose Generative Ensembles, a model-independent technique for OoD detection that combines density-based anomaly detection with uncertainty estimation. Our method outperforms ODIN and VIB baselines on image datasets, and achieves comparable performance to a classification model on the Kaggle Credit Fraud dataset. https://​github.com/​hschoi1/​rich_latent
 +https://​www.youtube.com/​watch?​v=2BpJcOf-1XA https://​github.com/​takashiishida/​pconf Binary Classification from Positive-Confidence Data