2021
Pfeifer, Anton; Lohweg, Volker
Classification of Faults in Cyber-Physical Systems with Complex-Valued Neural Networks Inproceedings Forthcoming
In: 26th IEEE International Conference on Emerging Technologies and Factory Automation (ETFA), IEEE, Forthcoming.
@inproceedings{PfeiferETFA2021,
title = {Classification of Faults in Cyber-Physical Systems with Complex-Valued Neural Networks},
author = {Anton Pfeifer and Volker Lohweg},
year = {2021},
date = {2021-09-07},
urldate = {2021-09-07},
booktitle = {26th IEEE International Conference on Emerging Technologies and Factory Automation (ETFA)},
publisher = {IEEE},
keywords = {},
pubstate = {forthcoming},
tppubtype = {inproceedings}
}
Tornede, Tanja; Tornede, Alexander; Wever, Marcel; Hüllermeier, Eyke
Coevolution of remaining useful lifetime estimation pipelines for automated predictive maintenance Inproceedings
In: Chicano, Francisco; Krawiec, Krzysztof (Hrsg.): GECCO '21: Genetic and Evolutionary Computation Conference, Lille, France, July 10-14, 2021, S. 368–376, ACM, 2021.
@inproceedings{DBLP:conf/gecco/TornedeTWH21,
title = {Coevolution of remaining useful lifetime estimation pipelines for automated predictive maintenance},
author = {Tanja Tornede and Alexander Tornede and Marcel Wever and Eyke Hüllermeier},
editor = {Francisco Chicano and Krzysztof Krawiec},
url = {https://doi.org/10.1145/3449639.3459395},
doi = {10.1145/3449639.3459395},
year = {2021},
date = {2021-06-26},
urldate = {2021-01-01},
booktitle = {GECCO '21: Genetic and Evolutionary Computation Conference, Lille,
France, July 10-14, 2021},
pages = {368--376},
publisher = {ACM},
abstract = {Automated machine learning (AutoML) strives for automatically constructing and configuring compositions of machine learning algorithms, called pipelines, with the goal to optimize a suitable performance measure on a concrete learning task. So far, most AutoML tools are focused on standard problem classes, such as classification and regression. In the field of predictive maintenance, especially the estimation of remaining useful lifetime (RUL), the task of AutoML becomes more complex. In particular, a good feature representation for multivariate sensor data is essential to achieve good performance. Due to the need for methods generating feature representations, the search space of candidate pipelines enlarges. Moreover, the runtime of a single pipeline increases substantially. In this paper, we tackle these problems by partitioning the search space into two sub-spaces, one for feature extraction methods and one for regression methods, and employ cooperative coevolution for searching a good combination. Thereby, we benefit from the fact that the generated feature representations can be cached, whence the evaluation of multiple regressors based on the same feature representation speeds up, allowing the evaluation of more candidate pipelines. Experimentally, we show that our coevolutionary strategy performs superior to the baselines.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Automated machine learning (AutoML) strives for automatically constructing and configuring compositions of machine learning algorithms, called pipelines, with the goal to optimize a suitable performance measure on a concrete learning task. So far, most AutoML tools are focused on standard problem classes, such as classification and regression. In the field of predictive maintenance, especially the estimation of remaining useful lifetime (RUL), the task of AutoML becomes more complex. In particular, a good feature representation for multivariate sensor data is essential to achieve good performance. Due to the need for methods generating feature representations, the search space of candidate pipelines enlarges. Moreover, the runtime of a single pipeline increases substantially. In this paper, we tackle these problems by partitioning the search space into two sub-spaces, one for feature extraction methods and one for regression methods, and employ cooperative coevolution for searching a good combination. Thereby, we benefit from the fact that the generated feature representations can be cached, whence the evaluation of multiple regressors based on the same feature representation speeds up, allowing the evaluation of more candidate pipelines. Experimentally, we show that our coevolutionary strategy performs superior to the baselines.
Shah, Zafran Hussain; Müller, Marcel; Wang, Tung-Cheng; Scheidig, Philip Maurice; Schneider, Axel; Schüttpelz, Mark; Huser, Thomas; Schenck, Wolfram
In: Photonics Research, 9 (5), S. B168–B181, 2021.
@article{shah2021deep,
title = {Deep-learning based denoising and reconstruction of super-resolution structured illumination microscopy images},
author = {Zafran Hussain Shah and Marcel Müller and Tung-Cheng Wang and Philip Maurice Scheidig and Axel Schneider and Mark Schüttpelz and Thomas Huser and Wolfram Schenck},
doi = {10.1364/PRJ.416437},
year = {2021},
date = {2021-04-14},
urldate = {2021-04-14},
journal = {Photonics Research},
volume = {9},
number = {5},
pages = {B168--B181},
publisher = {Optical Society of America},
abstract = {Super-resolution structured illumination microscopy (SR-SIM) provides an up to twofold enhanced spatial res- olution of fluorescently labeled samples. The reconstruction of high-quality SR-SIM images critically depends on patterned illumination with high modulation contrast. Noisy raw image data (e.g., as a result of low excitation power or low exposure time), result in reconstruction artifacts. Here, we demonstrate deep-learning based SR-SIM image denoising that results in high-quality reconstructed images. A residual encoding–decoding convolutional neural network (RED-Net) was used to successfully denoise computationally reconstructed noisy SR-SIM images. We also demonstrate the end-to-end deep-learning based denoising and reconstruction of raw SIM images into high-resolution SR-SIM images. Both image reconstruction methods prove to be very robust against image reconstruction artifacts and generalize very well across various noise levels. The combination of computational image reconstruction and subsequent denoising via RED-Net shows very robust performance during inference
after training even if the microscope settings change.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Super-resolution structured illumination microscopy (SR-SIM) provides an up to twofold enhanced spatial res- olution of fluorescently labeled samples. The reconstruction of high-quality SR-SIM images critically depends on patterned illumination with high modulation contrast. Noisy raw image data (e.g., as a result of low excitation power or low exposure time), result in reconstruction artifacts. Here, we demonstrate deep-learning based SR-SIM image denoising that results in high-quality reconstructed images. A residual encoding–decoding convolutional neural network (RED-Net) was used to successfully denoise computationally reconstructed noisy SR-SIM images. We also demonstrate the end-to-end deep-learning based denoising and reconstruction of raw SIM images into high-resolution SR-SIM images. Both image reconstruction methods prove to be very robust against image reconstruction artifacts and generalize very well across various noise levels. The combination of computational image reconstruction and subsequent denoising via RED-Net shows very robust performance during inference
after training even if the microscope settings change.
after training even if the microscope settings change.
Holst, Christoph-Alexander; Lohweg, Volker
A Redundancy Metric Set within Possibility Theory for Multi-Sensor Systems Artikel
In: Sensors, 21 (7), 2021.
@article{Article,
title = {A Redundancy Metric Set within Possibility Theory for Multi-Sensor Systems},
author = {Holst, Christoph-Alexander and Lohweg, Volker},
doi = {10.3390/s21072508},
year = {2021},
date = {2021-04-03},
urldate = {2021-01-01},
journal = {Sensors},
volume = {21},
number = {7},
abstract = {In intelligent technical multi-sensor systems, information is often at least partly redundant—either by design or inherently due to the dynamic processes of the observed system. If sensors are known to be redundant, (i) information processing can be engineered to be more robust against sensor failures, (ii) failures themselves can be detected more easily, and (iii) computational costs can be reduced. This contribution proposes a metric which quantifies the degree of redundancy between sensors. It is set within the possibility theory. Information coming from sensors in technical and cyber–physical systems are often imprecise, incomplete, biased, or affected by noise. Relations between information of sensors are often only spurious. In short, sensors are not fully reliable. The proposed metric adopts the ability of possibility theory to model incompleteness and imprecision exceptionally well. The focus is on avoiding the detection of spurious redundancy. This article defines redundancy in the context of possibilistic information, specifies requirements towards a redundancy metric, details the information processing, and evaluates the metric qualitatively on information coming from three technical datasets.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
In intelligent technical multi-sensor systems, information is often at least partly redundant—either by design or inherently due to the dynamic processes of the observed system. If sensors are known to be redundant, (i) information processing can be engineered to be more robust against sensor failures, (ii) failures themselves can be detected more easily, and (iii) computational costs can be reduced. This contribution proposes a metric which quantifies the degree of redundancy between sensors. It is set within the possibility theory. Information coming from sensors in technical and cyber–physical systems are often imprecise, incomplete, biased, or affected by noise. Relations between information of sensors are often only spurious. In short, sensors are not fully reliable. The proposed metric adopts the ability of possibility theory to model incompleteness and imprecision exceptionally well. The focus is on avoiding the detection of spurious redundancy. This article defines redundancy in the context of possibilistic information, specifies requirements towards a redundancy metric, details the information processing, and evaluates the metric qualitatively on information coming from three technical datasets.
Kummert, Johannes; Schulz, Alexander; Redick, Tim; Ayoub, Nassim; Modabber, Ali; Abel, Dirk; Hammer, Barbara
Efficient Reject Options for Particle Filter Object Tracking in Medical Applications Artikel
In: Sensors, 21 (6), S. 2114, 2021.
@article{kummert2021efficient,
title = {Efficient Reject Options for Particle Filter Object Tracking in Medical Applications},
author = {Johannes Kummert and Alexander Schulz and Tim Redick and Nassim Ayoub and Ali Modabber and Dirk Abel and Barbara Hammer},
doi = {https://doi.org/10.3390/s21062114},
year = {2021},
date = {2021-03-17},
urldate = {2021-03-17},
journal = {Sensors},
volume = {21},
number = {6},
pages = {2114},
publisher = {Multidisciplinary Digital Publishing Institute},
abstract = {Reliable object tracking that is based on video data constitutes an important challenge in diverse areas, including, among others, assisted surgery. Particle filtering offers a state-of-the-art technology for this challenge. Becaise a particle filter is based on a probabilistic model, it provides explicit likelihood values; in theory, the question of whether an object is reliably tracked can be addressed based on these values, provided that the estimates are correct. In this contribution, we investigate the question of whether these likelihood values are suitable for deciding whether the tracked object has been lost. An immediate strategy uses a simple threshold value to reject settings with a likelihood that is too small. We show in an application from the medical domain—object tracking in assisted surgery in the domain of Robotic Osteotomies—that this simple threshold strategy does not provide a reliable reject option for object tracking, in particular if different settings are considered. However, it is possible to develop reliable and flexible machine learning models that predict a reject based on diverse quantities that are computed by the particle filter. Modeling the task in the form of a regression enables a flexible handling of different demands on the tracking accuracy; modeling the challenge as an ensemble of classification tasks yet surpasses the results, while offering the same flexibility. },
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Reliable object tracking that is based on video data constitutes an important challenge in diverse areas, including, among others, assisted surgery. Particle filtering offers a state-of-the-art technology for this challenge. Becaise a particle filter is based on a probabilistic model, it provides explicit likelihood values; in theory, the question of whether an object is reliably tracked can be addressed based on these values, provided that the estimates are correct. In this contribution, we investigate the question of whether these likelihood values are suitable for deciding whether the tracked object has been lost. An immediate strategy uses a simple threshold value to reject settings with a likelihood that is too small. We show in an application from the medical domain—object tracking in assisted surgery in the domain of Robotic Osteotomies—that this simple threshold strategy does not provide a reliable reject option for object tracking, in particular if different settings are considered. However, it is possible to develop reliable and flexible machine learning models that predict a reject based on diverse quantities that are computed by the particle filter. Modeling the task in the form of a regression enables a flexible handling of different demands on the tracking accuracy; modeling the challenge as an ensemble of classification tasks yet surpasses the results, while offering the same flexibility.
2020
Holst, Christoph-Alexander; Lohweg, Volker
A Redundancy Metric based on the Framework of Possibility Theory for Technical Systems Inproceedings
In: 2020 IEEE 25th International Conference on Emerging Technologies and Factory Automation (ETFA), S. 1571-1578, IEEE, 2020.
@inproceedings{Holst.2020,
title = {A Redundancy Metric based on the Framework of Possibility Theory for Technical Systems},
author = {Christoph-Alexander Holst and Volker Lohweg},
doi = {10.1109/ETFA46521.2020.9212080},
year = {2020},
date = {2020-10-05},
urldate = {2020-01-01},
booktitle = {2020 IEEE 25th International Conference on Emerging Technologies and Factory Automation (ETFA)},
volume = {1},
pages = {1571-1578},
publisher = {IEEE},
abstract = {Detecting redundancies between information sources is essential for applications both in machine learning and information fusion. State-of-the-art redundancy metrics, such as correlation coefficients or mutual information, are based on probabilistic concepts. In technical multi-source systems informa- tion is often uncertain but also incomplete. Thus information is often provided with uncertainty distributions (probabilistic or possibilistic). In this paper a redundancy metric is proposed which is embedded in the framework of possibility theory applicable incomplete and uncertain information. The metric considers the consistency between sources, the specificity of pieces of information, and the range of observed information over the frame of discernment. The redundancy metric is designed to be cautious since incorrect identification of redundancies affects both machine learning and information fusion applications negatively. A machine learner may be deprived of informa- tion, whereas an information fusion system, relying on false assumptions, may, e. g. , incorrectly assess sources as unreliable. The proposed redundancy metric is qualitatively evaluated on information sources of three technical datasets.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Detecting redundancies between information sources is essential for applications both in machine learning and information fusion. State-of-the-art redundancy metrics, such as correlation coefficients or mutual information, are based on probabilistic concepts. In technical multi-source systems informa- tion is often uncertain but also incomplete. Thus information is often provided with uncertainty distributions (probabilistic or possibilistic). In this paper a redundancy metric is proposed which is embedded in the framework of possibility theory applicable incomplete and uncertain information. The metric considers the consistency between sources, the specificity of pieces of information, and the range of observed information over the frame of discernment. The redundancy metric is designed to be cautious since incorrect identification of redundancies affects both machine learning and information fusion applications negatively. A machine learner may be deprived of informa- tion, whereas an information fusion system, relying on false assumptions, may, e. g. , incorrectly assess sources as unreliable. The proposed redundancy metric is qualitatively evaluated on information sources of three technical datasets.
Tornede, Tanja; Tornede, Alexander; Wever, Marcel; Mohr, Felix; Hüllermeier, Eyke
AutoML for Predictive Maintenance: One Tool to RUL Them All Inproceedings
In: Gama, João; Pashami, Sepideh; Bifet, Albert; Mouchaweh, Moamar Sayed; Fröning, Holger; Pernkopf, Franz; Schiele, Gregor; Blott, Michaela (Hrsg.): IoT Streams for Data-Driven Predictive Maintenance and IoT, Edge, and Mobile for Embedded Machine Learning - Second International Workshop, IoT Streams 2020, and First International Workshop, ITEM 2020, Co-located with ECML/PKDD 2020, Ghent, Belgium, September 14-18, 2020, Revised Selected Papers, S. 106–118, Springer, 2020.
@inproceedings{DBLP:conf/pkdd/TornedeTWMH20,
title = {AutoML for Predictive Maintenance: One Tool to RUL Them All},
author = {Tanja Tornede and Alexander Tornede and Marcel Wever and Felix Mohr and Eyke Hüllermeier},
editor = {João Gama and Sepideh Pashami and Albert Bifet and Moamar Sayed Mouchaweh and Holger Fröning and Franz Pernkopf and Gregor Schiele and Michaela Blott},
url = {https://doi.org/10.1007/978-3-030-66770-2_8},
doi = {10.1007/978-3-030-66770-2_8},
year = {2020},
date = {2020-09-14},
urldate = {2020-01-01},
booktitle = {IoT Streams for Data-Driven Predictive Maintenance and IoT, Edge,
and Mobile for Embedded Machine Learning - Second International Workshop,
IoT Streams 2020, and First International Workshop, ITEM 2020, Co-located
with ECML/PKDD 2020, Ghent, Belgium, September 14-18, 2020, Revised
Selected Papers},
volume = {1325},
pages = {106--118},
publisher = {Springer},
series = {Communications in Computer and Information Science},
abstract = {Automated machine learning (AutoML) deals with the automatic composition and configuration of machine learning pipelines, including the selection and parametrization of preprocessors and learning algorithms. While recent work in this area has shown impressive results, existing approaches are essentially limited to standard problem classes such as classification and regression. In parallel, research in the field of predictive maintenance, particularly remaining useful lifetime (RUL) estimation, has received increasing attention, due to its practical relevance and potential to reduce unplanned downtime in industrial plants. However, applying existing AutoML methods to RUL estimation is non-trivial, as in this domain, one has to deal with varying-length multivariate time series data. Furthermore, the data often directly originates from real-world scenarios or simulations, and hence requires extensive preprocessing. In this work, we present ML-Plan-RUL, an adaptation of the AutoML tool ML-Plan to the problem of RUL estimation. To the best of our knowledge, it is the first tool specifically tailored towards automated RUL estimation, combining feature engineering, algorithm selection, and hyperparameter optimization into an end-to-end approach. First promising experimental results demonstrate the efficacy of ML-Plan-RUL.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Automated machine learning (AutoML) deals with the automatic composition and configuration of machine learning pipelines, including the selection and parametrization of preprocessors and learning algorithms. While recent work in this area has shown impressive results, existing approaches are essentially limited to standard problem classes such as classification and regression. In parallel, research in the field of predictive maintenance, particularly remaining useful lifetime (RUL) estimation, has received increasing attention, due to its practical relevance and potential to reduce unplanned downtime in industrial plants. However, applying existing AutoML methods to RUL estimation is non-trivial, as in this domain, one has to deal with varying-length multivariate time series data. Furthermore, the data often directly originates from real-world scenarios or simulations, and hence requires extensive preprocessing. In this work, we present ML-Plan-RUL, an adaptation of the AutoML tool ML-Plan to the problem of RUL estimation. To the best of our knowledge, it is the first tool specifically tailored towards automated RUL estimation, combining feature engineering, algorithm selection, and hyperparameter optimization into an end-to-end approach. First promising experimental results demonstrate the efficacy of ML-Plan-RUL.
Braun, Marco; Wrede, Sebastian
Incorporation of Expert Knowledge for Learning Robotic Assembly Tasks Inproceedings
In: 2020 IEEE 25th International Conference on Emerging Technologies and Factory Automation (ETFA), IEEE, 2020, ISBN: 978-1-7281-8956-7.
@inproceedings{Braun2020,
title = {Incorporation of Expert Knowledge for Learning Robotic Assembly Tasks},
author = {Marco Braun and Sebastian Wrede},
url = {https://ieeexplore.ieee.org/document/9211917/},
doi = {10.1109/ETFA46521.2020.9211917},
isbn = {978-1-7281-8956-7},
year = {2020},
date = {2020-09-08},
booktitle = {2020 IEEE 25th International Conference on Emerging Technologies and Factory Automation (ETFA)},
publisher = {IEEE},
abstract = {Autonomous learning of robotic manipulation tasks is a desirable proposition for the future of industrial manufacturing to increase flexibility and reduce manual engineering effort. In particular assembly tasks that require contact-rich manipulation skills are challenging to accomplish with classical robotic control methods. The Reinforcement Learning (RL) framework provides a possibility to learn complex behaviors based on interaction with the environment. Although a lot of research has been done robotic assembly tasks remain a challenge for pure learning-based systems. In this paper we give an overview on grey-box learning approaches that integrate prior knowledge and learning based methods. Different dimensions of knowledge injection are identified, and knowledge representations are described. These representations are discussed in the context of industrial assembly processes to answer the question: how can process experts model their knowledge to boost RL approaches in the context of industrial assembly?},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Autonomous learning of robotic manipulation tasks is a desirable proposition for the future of industrial manufacturing to increase flexibility and reduce manual engineering effort. In particular assembly tasks that require contact-rich manipulation skills are challenging to accomplish with classical robotic control methods. The Reinforcement Learning (RL) framework provides a possibility to learn complex behaviors based on interaction with the environment. Although a lot of research has been done robotic assembly tasks remain a challenge for pure learning-based systems. In this paper we give an overview on grey-box learning approaches that integrate prior knowledge and learning based methods. Different dimensions of knowledge injection are identified, and knowledge representations are described. These representations are discussed in the context of industrial assembly processes to answer the question: how can process experts model their knowledge to boost RL approaches in the context of industrial assembly?
Schulz, Alexander; Hinder, Fabian; Hammer, Barbara
In: Bessiere, Christian (Hrsg.): Proceedings of the Twenty-Ninth International Joint Conference on Artificial Intelligence, IJCAI-20, S. 2305–2311, International Joint Conferences on Artificial Intelligence Organization, 2020, (Main track).
@inproceedings{ijcai2020-319,
title = {DeepView: Visualizing Classification Boundaries of Deep Neural Networks as Scatter Plots Using Discriminative Dimensionality Reduction},
author = {Alexander Schulz and Fabian Hinder and Barbara Hammer},
editor = {Christian Bessiere},
url = {https://doi.org/10.24963/ijcai.2020/319},
doi = {10.24963/ijcai.2020/319},
year = {2020},
date = {2020-07-01},
urldate = {2020-01-01},
booktitle = {Proceedings of the Twenty-Ninth International Joint Conference on
Artificial Intelligence, IJCAI-20},
pages = {2305--2311},
publisher = {International Joint Conferences on Artificial Intelligence Organization},
abstract = {Machine learning algorithms using deep architectures have been able to implement increasingly powerful and successful models. However, they also become increasingly more complex, more difficult to comprehend and easier to fool. So far, most methods in the literature investigate the decision of the model for a single given input datum. In this paper, we propose to visualize a part of the decision function of a deep neural network together with a part of the data set in two dimensions with discriminative dimensionality reduction. This enables us to inspect how different properties of the data are treated by the model, such as outliers, adversaries or poisoned data. Further, the presented approach is complementary to the mentioned interpretation methods from the literature and hence might be even more useful in combination with those. Code is available at https://github.com/LucaHermes/DeepView},
note = {Main track},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Machine learning algorithms using deep architectures have been able to implement increasingly powerful and successful models. However, they also become increasingly more complex, more difficult to comprehend and easier to fool. So far, most methods in the literature investigate the decision of the model for a single given input datum. In this paper, we propose to visualize a part of the decision function of a deep neural network together with a part of the data set in two dimensions with discriminative dimensionality reduction. This enables us to inspect how different properties of the data are treated by the model, such as outliers, adversaries or poisoned data. Further, the presented approach is complementary to the mentioned interpretation methods from the literature and hence might be even more useful in combination with those. Code is available at https://github.com/LucaHermes/DeepView
Pfannschmidt, Lukas; Jakob, Jonathan; Hinder, Fabian; Biehl, Michael; Tino, Peter; Hammer, Barbara
In: Neurocomputing, 2020.
@article{pfannschmidt_feature_2019,
title = {Feature Relevance Determination for Ordinal Regression in the Context of Feature Redundancies and Privileged Information},
author = {Lukas Pfannschmidt and Jonathan Jakob and Fabian Hinder and Michael Biehl and Peter Tino and Barbara Hammer},
url = {https://www.sciencedirect.com/science/article/abs/pii/S0925231220305038},
doi = {10.1016/j.neucom.2019.12.133},
year = {2020},
date = {2020-04-09},
urldate = {2020-03-20},
journal = {Neurocomputing},
abstract = {Advances in machine learning technologies have led to increasingly powerful models in particular in the context of big data. Yet, many application scenarios demand for robustly interpretable models rather than optimum model accuracy; as an example, this is the case if potential biomarkers or causal factors should be discovered based on a set of given measurements. In this contribution, we focus on feature selection paradigms, which enable us to uncover relevant factors of a given regularity based on a sparse model. We focus on the important specific setting of linear ordinal regression, i.e. data have to be ranked into one of a finite number of ordered categories by a linear projection. Unlike previous work, we consider the case that features are potentially redundant, such that no unique minimum set of relevant features exists. We aim for an identification of all strongly and all weakly relevant features as well as their type of relevance (strong or weak); we achieve this goal by determining feature relevance bounds, which correspond to the minimum and maximum feature relevance, respectively, if searched over all equivalent models. In addition, we discuss how this setting enables us to substitute some of the features, e.g. due to their semantics, and how to extend the framework of feature relevance intervals to the setting of privileged information, i.e. potentially relevant information is available for training purposes only, but cannot be used for the prediction itself.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Advances in machine learning technologies have led to increasingly powerful models in particular in the context of big data. Yet, many application scenarios demand for robustly interpretable models rather than optimum model accuracy; as an example, this is the case if potential biomarkers or causal factors should be discovered based on a set of given measurements. In this contribution, we focus on feature selection paradigms, which enable us to uncover relevant factors of a given regularity based on a sparse model. We focus on the important specific setting of linear ordinal regression, i.e. data have to be ranked into one of a finite number of ordered categories by a linear projection. Unlike previous work, we consider the case that features are potentially redundant, such that no unique minimum set of relevant features exists. We aim for an identification of all strongly and all weakly relevant features as well as their type of relevance (strong or weak); we achieve this goal by determining feature relevance bounds, which correspond to the minimum and maximum feature relevance, respectively, if searched over all equivalent models. In addition, we discuss how this setting enables us to substitute some of the features, e.g. due to their semantics, and how to extend the framework of feature relevance intervals to the setting of privileged information, i.e. potentially relevant information is available for training purposes only, but cannot be used for the prediction itself.
Shaker, Mohammad Hossein; Hüllermeier, Eyke
Aleatoric and Epistemic Uncertainty with Random Forests Artikel
In: arXiv:2001.00893 [cs, stat], 2020, (arXiv: 2001.00893).
@article{shaker_aleatoric_2020,
title = {Aleatoric and Epistemic Uncertainty with Random Forests},
author = {Mohammad Hossein Shaker and Eyke Hüllermeier},
url = {http://arxiv.org/abs/2001.00893},
year = {2020},
date = {2020-01-03},
urldate = {2020-01-01},
journal = {arXiv:2001.00893 [cs, stat]},
abstract = {Due to the steadily increasing relevance of machine learning for practical applications, many of which are coming with safety requirements, the notion of uncertainty has received increasing attention in machine learning research in the last couple of years. In particular, the idea of distinguishing between two important types of uncertainty, often refereed to as aleatoric and epistemic, has recently been studied in the setting of supervised learning. In this paper, we propose to quantify these uncertainties with random forests. More specifically, we show how two general approaches for measuring the learnertextquoterights aleatoric and epistemic uncertainty in a prediction can be instantiated with decision trees and random forests as learning algorithms in a classification setting. In this regard, we also compare random forests with deep neural networks, which have been used for a similar purpose.},
note = {arXiv: 2001.00893},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Due to the steadily increasing relevance of machine learning for practical applications, many of which are coming with safety requirements, the notion of uncertainty has received increasing attention in machine learning research in the last couple of years. In particular, the idea of distinguishing between two important types of uncertainty, often refereed to as aleatoric and epistemic, has recently been studied in the setting of supervised learning. In this paper, we propose to quantify these uncertainties with random forests. More specifically, we show how two general approaches for measuring the learnertextquoterights aleatoric and epistemic uncertainty in a prediction can be instantiated with decision trees and random forests as learning algorithms in a classification setting. In this regard, we also compare random forests with deep neural networks, which have been used for a similar purpose.
2019
Holst, Christoph-Alexander; Lohweg, Volker
Feature fusion to increase the robustness of machine learners in industrial environments Artikel
In: at - Automatisierungstechnik, 67 (10), S. 853–865, 2019.
@article{holst2019feature,
title = {Feature fusion to increase the robustness of machine learners in industrial environments},
author = {Christoph-Alexander Holst and Volker Lohweg},
doi = {10.1515/auto-2019-0028},
year = {2019},
date = {2019-09-27},
urldate = {2019-01-01},
journal = {at - Automatisierungstechnik},
volume = {67},
number = {10},
pages = {853--865},
abstract = {Industrial applications put special demands on machine learning algorithms. Noisy data, outliers, and sensor faults present an immense challenge for learners. A considerable part of machine learning research focuses on the selection of relevant, non-redundant features. This contribution details an approach to group and fuse redundant features prior to learning and classification. Features are grouped relying on a correlation-based redundancy measure. The fusion of features is guided by determining the majority observation based on possibility distributions. Furthermore, this paper studies the effects of feature fusion on the robustness and performance of classification with a focus on industrial applications. The approach is statistically evaluated on public datasets in comparison to classification on selected features only.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Industrial applications put special demands on machine learning algorithms. Noisy data, outliers, and sensor faults present an immense challenge for learners. A considerable part of machine learning research focuses on the selection of relevant, non-redundant features. This contribution details an approach to group and fuse redundant features prior to learning and classification. Features are grouped relying on a correlation-based redundancy measure. The fusion of features is guided by determining the majority observation based on possibility distributions. Furthermore, this paper studies the effects of feature fusion on the robustness and performance of classification with a focus on industrial applications. The approach is statistically evaluated on public datasets in comparison to classification on selected features only.
Eiteneuer, Benedikt; Hranisavljevic, Nemanja; Niggemann, Oliver
Dimensionality Reduction and Anomaly Detection for CPPS Data using Autoencoder Artikel
In: International Conference on Industrial Technology, 2019.
@article{eiteneuer2019dimensionality,
title = {Dimensionality Reduction and Anomaly Detection for CPPS Data using Autoencoder},
author = {Benedikt Eiteneuer and Nemanja Hranisavljevic and Oliver Niggemann},
url = {https://ieeexplore.ieee.org/document/8755116},
doi = {10.1109/ICIT.2019.8755116},
year = {2019},
date = {2019-07-04},
urldate = {2019-01-01},
journal = {International Conference on Industrial Technology},
abstract = {Unsupervised anomaly detection (AD) is a major topic in the field of Cyber-Physical Production Systems (CPPSs). A closely related concern is dimensionality reduction (DR) which is: 1) often used as a preprocessing step in an AD solution, 2) a sort of AD, if a measure of observation conformity to the learned data manifold is provided. We argue that the two aspects can be complementary in a CPPS anomaly detection solution. In this work, we focus on the nonlinear autoencoder (AE) as a DR/AD approach. The contribution of this work is: 1) we examine the suitability of AE reconstruction error as an AD decision criterion in CPPS data. 2) we analyze its relation to a potential second-phase AD approach in the AE latent space 3) we evaluate the performance of the approach on three real-world datasets. Moreover, the approach outperforms state-of-the-art techniques, alongside a relatively simple and straightforward application.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Unsupervised anomaly detection (AD) is a major topic in the field of Cyber-Physical Production Systems (CPPSs). A closely related concern is dimensionality reduction (DR) which is: 1) often used as a preprocessing step in an AD solution, 2) a sort of AD, if a measure of observation conformity to the learned data manifold is provided. We argue that the two aspects can be complementary in a CPPS anomaly detection solution. In this work, we focus on the nonlinear autoencoder (AE) as a DR/AD approach. The contribution of this work is: 1) we examine the suitability of AE reconstruction error as an AD decision criterion in CPPS data. 2) we analyze its relation to a potential second-phase AD approach in the AE latent space 3) we evaluate the performance of the approach on three real-world datasets. Moreover, the approach outperforms state-of-the-art techniques, alongside a relatively simple and straightforward application.
Göpfert, Christina; Ben-David, Shai; Bousquet, Olivier; Gelly, Sylvain; Tolstikhin, Ilya; Urner, Ruth
When can unlabeled data improve the learning rate? Artikel
In: arXiv preprint arXiv:1905.11866, 2019.
@article{gopfert2019can,
title = {When can unlabeled data improve the learning rate?},
author = {Christina Göpfert and Shai Ben-David and Olivier Bousquet and Sylvain Gelly and Ilya Tolstikhin and Ruth Urner},
url = {https://arxiv.org/abs/1905.11866},
year = {2019},
date = {2019-06-25},
urldate = {2019-01-01},
journal = {arXiv preprint arXiv:1905.11866},
abstract = {In semi-supervised classification, one is given access both to labeled and unlabeled data. As unlabeled data is typically cheaper to acquire than labeled data, this setup becomes advantageous as soon as one can exploit the unlabeled data in order to produce a better classifier than with labeled data alone. However, the conditions under which such an improvement is possible are not fully understood yet. Our analysis focuses on improvements in the minimax learning rate in terms of the number of labeled examples (with the number of unlabeled examples being allowed to depend on the number of labeled ones). We argue that for such improvements to be realistic and indisputable, certain specific conditions should be satisfied and previous analyses have failed to meet those conditions. We then demonstrate examples where these conditions can be met, in particular showing rate changes from 1/√ℓ to e^−cℓ and from 1/√ℓ to 1/ℓ. These results improve our understanding of what is and isn't possible in semi-supervised learning.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
In semi-supervised classification, one is given access both to labeled and unlabeled data. As unlabeled data is typically cheaper to acquire than labeled data, this setup becomes advantageous as soon as one can exploit the unlabeled data in order to produce a better classifier than with labeled data alone. However, the conditions under which such an improvement is possible are not fully understood yet. Our analysis focuses on improvements in the minimax learning rate in terms of the number of labeled examples (with the number of unlabeled examples being allowed to depend on the number of labeled ones). We argue that for such improvements to be realistic and indisputable, certain specific conditions should be satisfied and previous analyses have failed to meet those conditions. We then demonstrate examples where these conditions can be met, in particular showing rate changes from 1/√ℓ to e^−cℓ and from 1/√ℓ to 1/ℓ. These results improve our understanding of what is and isn't possible in semi-supervised learning.
Bifet, Albert; Hammer, Barbara; Schleif, Frank-Michael
Recent trends in streaming data analysis, concept drift and analysis of dynamic data sets Artikel
In: Proc. European Symposium on Artificial Neural Networks, 2019.
@article{bifet2019streaming,
title = {Recent trends in streaming data analysis, concept drift and analysis of dynamic data sets},
author = {Albert Bifet and Barbara Hammer and Frank-Michael Schleif},
url = {https://www.elen.ucl.ac.be/Proceedings/esann/esannpdf/es2019-3.pdf},
year = {2019},
date = {2019-04-24},
urldate = {2019-01-01},
journal = {Proc. European Symposium on Artificial Neural Networks},
abstract = {Today, many data are not any longer static but occur as dynamic data streams with high velocity, variability and volume. This leads to new challenges to be addressed by novel or adapted algorithms. In this tutorial we provide an introduction into the field of streaming data analysis summarizing its major characteristics and highlighting important research directions in the analysis of dynamic data.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Today, many data are not any longer static but occur as dynamic data streams with high velocity, variability and volume. This leads to new challenges to be addressed by novel or adapted algorithms. In this tutorial we provide an introduction into the field of streaming data analysis summarizing its major characteristics and highlighting important research directions in the analysis of dynamic data.
Pfannschmidt, Lukas; Jakob, Jonathan; Biehl, Michael; Tino, Peter; Hammer, Barbara
Feature Relevance Bounds for Ordinal Regression Artikel
In: Proc. European Symposium on Artificial Neural Networks, 2019., 2019.
@article{pfannschmidt2019feature,
title = {Feature Relevance Bounds for Ordinal Regression},
author = {Lukas Pfannschmidt and Jonathan Jakob and Michael Biehl and Peter Tino and Barbara Hammer},
url = {https://www.elen.ucl.ac.be/Proceedings/esann/esannpdf/es2019-162.pdf},
year = {2019},
date = {2019-02-20},
urldate = {2019-01-01},
journal = {Proc. European Symposium on Artificial Neural Networks, 2019.},
abstract = {The increasing occurrence of ordinal data, mainly sociodemo- graphic, led to a renewed research interest in ordinal regression, i.e. the prediction of ordered classes. Besides model accuracy, the interpretation of these models itself is of high relevance, and existing approaches therefore enforce e.g. model sparsity. For high dimensional or highly correlated data, however, this might be misleading due to strong variable dependencies. In this contribution, we aim for an identification of feature relevance bounds which – besides identifying all relevant features – explicitly differentiates between strongly and weakly relevant features.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
The increasing occurrence of ordinal data, mainly sociodemo- graphic, led to a renewed research interest in ordinal regression, i.e. the prediction of ordered classes. Besides model accuracy, the interpretation of these models itself is of high relevance, and existing approaches therefore enforce e.g. model sparsity. For high dimensional or highly correlated data, however, this might be misleading due to strong variable dependencies. In this contribution, we aim for an identification of feature relevance bounds which – besides identifying all relevant features – explicitly differentiates between strongly and weakly relevant features.
Brinkrolf, Johannes; Hammer, Barbara
Time integration and reject options for probabilistic output of pairwise LVQ Artikel
In: Neural Computing and Applications, S. 1–14, 2019.
@article{brinkrolf2019time,
title = {Time integration and reject options for probabilistic output of pairwise LVQ},
author = {Johannes Brinkrolf and Barbara Hammer},
url = {https://link.springer.com/content/pdf/10.1007/s00521-018-03966-0.pdf},
doi = {10.1007/s00521-018-03966-0},
year = {2019},
date = {2019-01-05},
urldate = {2019-01-01},
journal = {Neural Computing and Applications},
pages = {1--14},
publisher = {Springer},
abstract = {Learning vector quantization (LVQ) constitutes a very popular machine learning technology with applications, for example, in biomedical data analysis, predictive maintenance/quality as well as product individualization. Albeit proba- bilistic LVQ variants exist, its deterministic counterparts are often preferred due to their better efficiency. The latter do not allow an immediate probabilistic interpretation of its output; hence, a rejection of classification based on confidence values is not possible. In this contribution, we investigate different schemes how to extend and integrate pairwise LVQ schemes to an overall probabilistic output, in comparison with a recent heuristic surrogate measure for the security of the classification, which is directly based on LVQ’s multi-class classification scheme. Furthermore, we propose a canonic way how to fuse these values over a given time window in case a possibly disrupted measurement is taken over a longer time interval to counter the uncertainty of a single point in time. Experimental results indicate that an explicit probabilistic treatment often yields superior results as compared to a standard deterministic LVQ method, but metric learning is able to annul this difference. Fusion over a short time period is beneficial in case of an unclear classification.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Learning vector quantization (LVQ) constitutes a very popular machine learning technology with applications, for example, in biomedical data analysis, predictive maintenance/quality as well as product individualization. Albeit proba- bilistic LVQ variants exist, its deterministic counterparts are often preferred due to their better efficiency. The latter do not allow an immediate probabilistic interpretation of its output; hence, a rejection of classification based on confidence values is not possible. In this contribution, we investigate different schemes how to extend and integrate pairwise LVQ schemes to an overall probabilistic output, in comparison with a recent heuristic surrogate measure for the security of the classification, which is directly based on LVQ’s multi-class classification scheme. Furthermore, we propose a canonic way how to fuse these values over a given time window in case a possibly disrupted measurement is taken over a longer time interval to counter the uncertainty of a single point in time. Experimental results indicate that an explicit probabilistic treatment often yields superior results as compared to a standard deterministic LVQ method, but metric learning is able to annul this difference. Fusion over a short time period is beneficial in case of an unclear classification.