Anomaly detection plays a pivotal role in diverse realworld applications such as cybersecurity, fault detection, network
nitoring, predictive maintenance, and highly automated driving. However, obtaining labeled anomalous data can be a formidable
allenge, especially when anomalies exhibit temporal evolution. This paper introduces LATAM (Long short-term memory Autoencoder with Temporal Attention Mechanism) for few-shot anomaly detection, with the aim of enhancing detection performance in scenarios with limited labeled anomaly data. LATAM effectively captures temporal dependencies and emphasizes significant patterns in multivariate time series data. In our investigation, we
mprehensively evaluate LATAM against other anomaly detection models, particularly assessing its capability in few-shot learning
enarios where we have minimal examples from the normal class and none from the anomalous class in the training data. Our
perimental results, derived from real-world photovoltaic inverter data, highlight LATAM’s superiority, showcasing a substantial
% mean F1 score improvement, even when trained on a mere two-week dataset. Furthermore, LATAM demonstrates remarkable
sults on the open-source SWaT dataset, achieving a 12% boost in accuracy with only two days of training data. Moreover, we
troduce a simple yet effective dynamic thresholding mechanism, further enhancing the anomaly detection capabilities of LATAM.
is underscores LATAM’s efficacy in addressing the challenges posed by limited labeled anomalies in practical scenarios and it
oves valuable for downstream tasks involving temporal representation and time series prediction, extending its utility beyond
omaly detection applications.

@inproceedings{nivarthi2023towards, author = {Nivarthi, Chandana Priya and Sick, Bernhard}, title = {Towards Few-Shot Time Series Anomaly Detection with Temporal Attention and Dynamic Thresholding}, booktitle = {International Conference on Machine Learning and Applications (ICMLA)}, publisher = {IEEE}, year = {2023}, pages = {1444--1450}, doi = {10.1109/ICMLA58977.2023.00218}, keywords = {imported, itegpub, isac-www, few-shot, learning, anomaly, detection, temporal, attention, dynamic, thresholding}, abstract = {Anomaly detection plays a pivotal role in diverse realworld applications such as cybersecurity, fault detection, network
nitoring, predictive maintenance, and highly automated driving. However, obtaining labeled anomalous data can be a formidable
allenge, especially when anomalies exhibit temporal evolution. This paper introduces LATAM (Long short-term memory Autoencoder with Temporal Attention Mechanism) for few-shot anomaly detection, with the aim of enhancing detection performance in scenarios with limited labeled anomaly data. LATAM effectively captures temporal dependencies and emphasizes significant patterns in multivariate time series data. In our investigation, we
mprehensively evaluate LATAM against other anomaly detection models, particularly assessing its capability in few-shot learning
enarios where we have minimal examples from the normal class and none from the anomalous class in the training data. Our
perimental results, derived from real-world photovoltaic inverter data, highlight LATAM’s superiority, showcasing a substantial
% mean F1 score improvement, even when trained on a mere two-week dataset. Furthermore, LATAM demonstrates remarkable
sults on the open-source SWaT dataset, achieving a 12% boost in accuracy with only two days of training data. Moreover, we
troduce a simple yet effective dynamic thresholding mechanism, further enhancing the anomaly detection capabilities of LATAM.
is underscores LATAM’s efficacy in addressing the challenges posed by limited labeled anomalies in practical scenarios and it
oves valuable for downstream tasks involving temporal representation and time series prediction, extending its utility beyond
omaly detection applications.} }

%0 = inproceedings %A = Nivarthi, Chandana Priya and Sick, Bernhard %B = International Conference on Machine Learning and Applications (ICMLA) %D = 2023 %I = IEEE %T = Towards Few-Shot Time Series Anomaly Detection with Temporal Attention and Dynamic Thresholding

@inproceedings{mitchell2015, author = {Mitchell, T. and Cohen, W. and Hruscha, E. and Talukdar, P. and Betteridge, J. and Carlson, A. and Dalvi, B. and Gardner, M. and Kisiel, B. and Krishnamurthy, J. and Lao, N. and Mazaitis, K. and Mohammad, T. and Nakashole, N. and Platanios, E. and Ritter, A. and Samadi, M. and Settles, B. and Wang, R. and Wijaya, D. and Gupta, A. and Chen, X. and Saparov, A. and Greaves, M. and Welling, J.}, title = {Never-Ending Learning}, booktitle = {AAAI}, year = {2015}, note = {: Never-Ending Learning in AAAI-2015}, url = {http://www.cs.cmu.edu/~wcohen/pubs.html}, keywords = {learning, nell, ontology, semantic, toread} }

%0 = inproceedings %A = Mitchell, T. and Cohen, W. and Hruscha, E. and Talukdar, P. and Betteridge, J. and Carlson, A. and Dalvi, B. and Gardner, M. and Kisiel, B. and Krishnamurthy, J. and Lao, N. and Mazaitis, K. and Mohammad, T. and Nakashole, N. and Platanios, E. and Ritter, A. and Samadi, M. and Settles, B. and Wang, R. and Wijaya, D. and Gupta, A. and Chen, X. and Saparov, A. and Greaves, M. and Welling, J. %B = AAAI %D = 2015 %T = Never-Ending Learning %U = http://www.cs.cmu.edu/~wcohen/pubs.html

@article{mnih2015humanlevel, author = {Mnih, Volodymyr and Kavukcuoglu, Koray and Silver, David and Rusu, Andrei A. and Veness, Joel and Bellemare, Marc G. and Graves, Alex and Riedmiller, Martin and Fidjeland, Andreas K. and Ostrovski, Georg and Petersen, Stig and Beattie, Charles and Sadik, Amir and Antonoglou, Ioannis and King, Helen and Kumaran, Dharshan and Wierstra, Daan and Legg, Shane and Hassabis, Demis}, title = {Human-level control through deep reinforcement learning}, journal = {Nature}, publisher = {Nature Publishing Group, a division of Macmillan Publishers Limited. All Rights Reserved.}, year = {2015}, volume = {518}, number = {7540}, pages = {529--533}, url = {http://dx.doi.org/10.1038/nature14236}, issn = {00280836}, keywords = {deep, learning, toread} }

%0 = article %A = Mnih, Volodymyr and Kavukcuoglu, Koray and Silver, David and Rusu, Andrei A. and Veness, Joel and Bellemare, Marc G. and Graves, Alex and Riedmiller, Martin and Fidjeland, Andreas K. and Ostrovski, Georg and Petersen, Stig and Beattie, Charles and Sadik, Amir and Antonoglou, Ioannis and King, Helen and Kumaran, Dharshan and Wierstra, Daan and Legg, Shane and Hassabis, Demis %D = 2015 %I = Nature Publishing Group, a division of Macmillan Publishers Limited. All Rights Reserved. %T = Human-level control through deep reinforcement learning %U = http://dx.doi.org/10.1038/nature14236

@inproceedings{ring2015condist, author = {Ring, Markus and Otto, Florian and Becker, Martin and Niebler, Thomas and Landes, Dieter and Hotho, Andreas}, title = {ConDist: A Context-Driven Categorical Distance Measure}, editor = {ECMLPKDD2015}, year = {2015}, keywords = {2015, categorical, data, learning, measure, myown, similarity, unsupervised} }

%0 = inproceedings %A = Ring, Markus and Otto, Florian and Becker, Martin and Niebler, Thomas and Landes, Dieter and Hotho, Andreas %D = 2015 %T = ConDist: A Context-Driven Categorical Distance Measure

@inproceedings{DBLP:conf/dsaa/KrompassNT14, author = {Krompass, Denis and Nickel, Maximilian and Tresp, Volker}, title = {Large-scale factorization of type-constrained multi-relational data}, booktitle = {International Conference on Data Science and Advanced Analytics, DSAA 2014, Shanghai, China, October 30 - November 1, 2014}, publisher = {IEEE}, year = {2014}, pages = {18--24}, url = {http://dx.doi.org/10.1109/DSAA.2014.7058046}, doi = {10.1109/DSAA.2014.7058046}, isbn = {978-1-4799-6991-3}, keywords = {graph, knowledge, learning, toread} }

%0 = inproceedings %A = Krompass, Denis and Nickel, Maximilian and Tresp, Volker %B = International Conference on Data Science and Advanced Analytics, DSAA 2014, Shanghai, China, October 30 - November 1, 2014 %D = 2014 %I = IEEE %T = Large-scale factorization of type-constrained multi-relational data %U = http://dx.doi.org/10.1109/DSAA.2014.7058046

@incollection{pol_introduction, author = {Lehmann, Jens and Voelker, Johanna}, title = {An Introduction to Ontology Learning}, editor = {Lehmann, Jens and Voelker, Johanna}, booktitle = {Perspectives on Ontology Learning}, publisher = {AKA / IOS Press}, year = {2014}, pages = {ix-xvi}, url = {http://jens-lehmann.org/files/2014/pol_introduction.pdf}, keywords = {introduction, learning, ontology} }

%0 = incollection %A = Lehmann, Jens and Voelker, Johanna %B = Perspectives on Ontology Learning %D = 2014 %I = AKA / IOS Press %T = An Introduction to Ontology Learning %U = http://jens-lehmann.org/files/2014/pol_introduction.pdf

Conditional Random Fields (CRF) are popular methods for labeling unstructured or textual data. Like many machine learning approaches, these undirected graphical models assume the instances to be independently distributed. However, in real-world applications data is grouped in a natural way, e.g., by its creation context. The instances in each group often share additional structural consistencies. This paper proposes a domain-independent method for exploiting these consistencies by combining two CRFs in a stacked learning framework. We apply rule learning collectively on the predictions of an initial CRF for one context to acquire descriptions of its specific properties. Then, we utilize these descriptions as dynamic and high quality features in an additional (stacked) CRF. The presented approach is evaluated with a real-world dataset for the segmentation of references and achieves a significant reduction of the labeling error.

@article{kluegl2013exploiting, author = {Kluegl, Peter and Toepfer, Martin and Lemmerich, Florian and Hotho, Andreas and Puppe, Frank}, title = {Exploiting Structural Consistencies with Stacked Conditional Random Fields}, journal = {Mathematical Methodologies in Pattern Recognition and Machine Learning Springer Proceedings in Mathematics & Statistics}, year = {2013}, volume = {30}, pages = {111-125}, keywords = {2013, ie, learning, myown, references}, abstract = {Conditional Random Fields (CRF) are popular methods for labeling unstructured or textual data. Like many machine learning approaches, these undirected graphical models assume the instances to be independently distributed. However, in real-world applications data is grouped in a natural way, e.g., by its creation context. The instances in each group often share additional structural consistencies. This paper proposes a domain-independent method for exploiting these consistencies by combining two CRFs in a stacked learning framework. We apply rule learning collectively on the predictions of an initial CRF for one context to acquire descriptions of its specific properties. Then, we utilize these descriptions as dynamic and high quality features in an additional (stacked) CRF. The presented approach is evaluated with a real-world dataset for the segmentation of references and achieves a significant reduction of the labeling error.} }

%0 = article %A = Kluegl, Peter and Toepfer, Martin and Lemmerich, Florian and Hotho, Andreas and Puppe, Frank %D = 2013 %T = Exploiting Structural Consistencies with Stacked Conditional Random Fields

The multi-label classification problem has generated significant interest in
cent years. However, existing approaches do not adequately address two key
allenges: (a) the ability to tackle problems with a large number (say
llions) of labels, and (b) the ability to handle data with missing labels. In
is paper, we directly address both these problems by studying the multi-label
oblem in a generic empirical risk minimization (ERM) framework. Our
amework, despite being simple, is surprisingly able to encompass several
cent label-compression based methods which can be derived as special cases of
r method. To optimize the ERM problem, we develop techniques that exploit the
ructure of specific loss functions - such as the squared loss function - to
fer efficient algorithms. We further show that our learning framework admits
rmal excess risk bounds even in the presence of missing labels. Our risk
unds are tight and demonstrate better generalization performance for low-rank
omoting trace-norm regularization when compared to (rank insensitive)
obenius norm regularization. Finally, we present extensive empirical results
a variety of benchmark datasets and show that our methods perform
gnificantly better than existing label compression based methods and can
ale up to very large datasets such as the Wikipedia dataset.

@misc{yu2013largescale, author = {Yu, Hsiang-Fu and Jain, Prateek and Kar, Purushottam and Dhillon, Inderjit S.}, title = {Large-scale Multi-label Learning with Missing Labels}, year = {2013}, note = {cite arxiv:1307.5101}, url = {http://arxiv.org/abs/1307.5101}, keywords = {classification, kallimachos, label, large, learning, multi}, abstract = {The multi-label classification problem has generated significant interest inrecent years. However, existing approaches do not adequately address two keychallenges: (a) the ability to tackle problems with a large number (saymillions) of labels, and (b) the ability to handle data with missing labels. Inthis paper, we directly address both these problems by studying the multi-labelproblem in a generic empirical risk minimization (ERM) framework. Ourframework, despite being simple, is surprisingly able to encompass severalrecent label-compression based methods which can be derived as special cases ofour method. To optimize the ERM problem, we develop techniques that exploit thestructure of specific loss functions - such as the squared loss function - tooffer efficient algorithms. We further show that our learning framework admitsformal excess risk bounds even in the presence of missing labels. Our riskbounds are tight and demonstrate better generalization performance for low-rankpromoting trace-norm regularization when compared to (rank insensitive)Frobenius norm regularization. Finally, we present extensive empirical resultson a variety of benchmark datasets and show that our methods performsignificantly better than existing label compression based methods and canscale up to very large datasets such as the Wikipedia dataset.} }

%0 = misc %A = Yu, Hsiang-Fu and Jain, Prateek and Kar, Purushottam and Dhillon, Inderjit S. %B = } %C = %D = 2013 %I = %T = Large-scale Multi-label Learning with Missing Labels} %U = http://arxiv.org/abs/1307.5101

@inproceedings{noauthororeditor, author = {Mirowski, Piotr and Ranzato, Marc'Aurelio and LeCun, Yann}, title = {Dynamic Auto-Encoders for Semantic Indexing}, editor = {of the NIPS 2010 Workshop on Deep Learning, Proceedings}, year = {2010}, url = {http://yann.lecun.com/exdb/publis/pdf/mirowski-nipsdl-10.pdf}, keywords = {deep, kallimachos, lda, learning, model, toread} }

%0 = inproceedings %A = Mirowski, Piotr and Ranzato, Marc'Aurelio and LeCun, Yann %D = 2010 %T = Dynamic Auto-Encoders for Semantic Indexing %U = http://yann.lecun.com/exdb/publis/pdf/mirowski-nipsdl-10.pdf

Random forests are a combination of tree predictors such that each tree depends on the values of a random vector sampled independently and with the same distribution for all trees in the forest. The generalization error for forests converges a.s. to a limit as the number of trees in the forest becomes large. The generalization error of a forest of tree classifiers depends on the strength of the individual trees in the forest and the correlation between them. Using a random selection of features to split each node yields error rates that compare favorably to

@article{breiman2001random, author = {Breiman, Leo}, title = {Random Forests}, journal = {Machine Learning}, publisher = {Kluwer Academic Publishers}, year = {2001}, volume = {45}, number = {1}, pages = {5-32}, url = {http://dx.doi.org/10.1023/A%3A1010933404324}, doi = {10.1023/A:1010933404324}, issn = {0885-6125}, keywords = {classification, ensemble, forest, learning, random}, abstract = {Random forests are a combination of tree predictors such that each tree depends on the values of a random vector sampled independently and with the same distribution for all trees in the forest. The generalization error for forests converges a.s. to a limit as the number of trees in the forest becomes large. The generalization error of a forest of tree classifiers depends on the strength of the individual trees in the forest and the correlation between them. Using a random selection of features to split each node yields error rates that compare favorably to } }

%0 = article %A = Breiman, Leo %D = 2001 %I = Kluwer Academic Publishers %T = Random Forests %U = http://dx.doi.org/10.1023/A%3A1010933404324