101 |
Learning Vector Symbolic Architectures for Reactive Robot BehavioursNeubert, Peer, Schubert, Stefan, Protzel, Peter 08 August 2017 (has links)
Vector Symbolic Architectures (VSA) combine a hypervector space and a set of operations on these vectors. Hypervectors provide powerful and noise-robust representations and VSAs are associated with promising theoretical properties for approaching high-level cognitive tasks. However, a major drawback of VSAs is the lack of opportunities to learn them from training data. Their power is merely an effect of good (and elaborate) design rather than learning. We exploit high-level knowledge about the structure of reactive robot problems to learn a VSA based on training data. We demonstrate preliminary results on a simple navigation task. Given a successful demonstration of a navigation run by pairs of sensor input and actuator output, the system learns a single hypervector that encodes this reactive behaviour. When executing (and combining) such VSA-based behaviours, the advantages of hypervectors (i.e. the representational power and robustness to noise) are preserved. Moreover, a particular beauty of this approach is that it can learn encodings for behaviours that have exactly the same form (a hypervector) no matter how complex the sensor input or the behaviours are.
|
102 |
Quantifying Dislocation MicrostructuresSteinberger, Dominik 28 August 2020 (has links)
In this work, we reconstructed and full characterized a dislocation microstructure that formed during an in situ micro-cantilever beam experiment. Based on this information, we were then able to infer how the dislocations propagated from the notch into the specimen.
We propose using the so-called 'discrete-to-continuous' (D2C) method, which converts discrete dislocation data to continuum fields, as a means to quantify microstructures. With this method, we studied how different methods of initializing the microstructure in discrete dislocation dynamics simulations affects the resulting microstructure. We found that not considering cross-slip leads to very different microstructures, and that cross-slip results in more similar microstructures.
Further, we used the continuum fields extracted via the D2C methods as input features for machine learning models for the classification of dislocation microstructures in nanoparticles. We found them to be well suited and that the combination of continuum fields is dependent on whether the microstructure is dominated by statistically stored or geometrically necessary dislocations.
|
103 |
A Revision of Procedural Knowledge in the conML FrameworkGroße, Florian Peter 24 March 2022 (has links)
Machine learning methods have been used very successfully for quite some time to recognize patterns, model correlations and generate hypotheses. However, the possibilities for weighing and evaluating the resulting models and hypotheses, and the search for alternatives and contradictions are still predominantly reserved for humans.
For this purpose, the novel concept of constructivist machine learning (conML) formalizes limitations of model validity and employs constructivist learning theory to enable doubting of new and existing models with the possibility of integrating, discarding, combining, and abstracting knowledge.
The present work identifies issues that impede the systems capability to abstract knowledge from generated models for tasks that lie in the domain of procedural knowledge, and proposes and implements identified solutions. To this end, the conML framework has been reimplemented in the Julia programming language and subsequently been extended.
Using a synthetic dataset of impedance spectra of modeled epithelia that has previously been analyzed with an existing implementation of conML, existing and new implementations are tested for consistency and proposed algorithmic changes are evaluated with respect to changes in model generation and abstraction ability when exploring unknown data.
Recommendations for specific settings and suggestions for further research are derived from the results. In terms of performance, flexibility and extensibility, the new implementation of conML in Julia provides a good starting point for further research and application of the system.:Contents
Abstract . . . . . III
Zusammenfassung . . . . . IV
Danksagung . . . . . V
Selbstständigkeitserklärung . . . . . V
1. Introduction
1.1. Research Questions . . . . . 2
2. Related Work
2.1. Hybrid AI Systems . . . . . 5
2.2. Constructivist Machine Learning (conML) . . . . . 6
2.3. Implemented Methods . . . . . 9
2.3.1. Unsupervised Machine Learning . . . . . 9
2.3.2. Supervised Machine Learning . . . . . 11
2.3.3. Supervised Feature Selection . . . . . 13
2.3.4. Unsupervised Feature Selection . . . . . 17
3. Methods and Implementation
3.1. Notable Algorithmic Changes . . . . . 19
3.1.1. Rescaling of Target Values . . . . . 19
3.1.2. ExtendedWinner Selection . . . . . 21
3.2. Package Structure . . . . . 23
3.3. Interfaces and Implementation of Specific Methods . . . . . 29
3.4. Datasets . . . . . 41
4. Results
4.1. Validation Against the conML Prototype . . . . . 43
4.2. Change in Abstraction Capability . . . . . 49
4.2.1. Influence of Target Scaling . . . . . 49
4.2.2. Influence of the Parameter kappa_p . . . . . 55
4.2.3. Influence of the Winner Selection Procedure . . . . . 61
5. Discussion
5.1. Reproduction Results . . . . . 67
5.2. Rescaling of Constructed Targets . . . . . 69
5.3. kappa_p and the Selection of Winner Models . . . . . 71
6. Conclusions
6.1. Contributions of this Work . . . . . 77
6.2. Future Work . . . . . 78
A. Julia Language Reference . . . . . 81
B. Additional Code Listings . . . . . 91
C. Available Parameters . . . . . 99
C.1. Block Processing . . . . . 105
D. Configurations Reference . . . . . 107
D.1. Unsupervised Methods . . . . . 107
D.2. Supervised Methods . . . . . 108
D.3. Feature Selection . . . . . 109
D.4. Winner Selection . . . . . 110
D.5. General Settings . . . . . 110
E. Supplemental Figures . . . . . 113
E.1. Replacing MAPE with RMSE for Z-Transform Target Scaling . . . . . 113
E.2. Combining Target Rescaling, Winner Selection and High kappa_p . . . . . 119
Bibliography . . . . . 123
List of Figures . . . . . 129
List of Listings . . . . . 133
List of Tables . . . . . 135 / Maschinelle Lernverfahren werden seit geraumer Zeit sehr erfolgreich zum Erkennen von Mustern, Abbilden von Zusammenhängen und Generieren von Hypothesen eingesetzt. Die Möglichkeiten zum Abwägen und Bewerten der entstandenen Modelle und Hypothesen, und die Suche nach Alternativen und Widersprüchen sind jedoch noch überwiegend dem Menschen vorbehalten.
Das neuartige Konzept des konstruktivistischen maschinellen Lernens (conML) formalisiert dazu die Grenzen der Gültigkeit von Modellen und ermöglicht mittels konstruktivistischer Lerntheorie ein Zweifeln über neue und bestehende Modelle mit der Möglichkeit zum Integrieren, Verwerfen, Kombinieren und Abstrahieren von Wissen.
Die vorliegende Arbeit identifiziert Probleme, die die Abstraktionsfähigkeit des Systems bei Aufgabenstellungen in der Prozeduralen Wissensdomäne einschränken, bietet Lösungsvorschläge und beschreibt deren Umsetzung. Das algorithmische Framework conML ist dazu in der Programmiersprache Julia reimplementiert und anschließend erweitert worden.
Anhand eines synthetischen Datensatzes von Impedanzspektren modellierter Epithelien, der bereits mit einem Prototypen des conML Systems analysiert worden ist, werden bestehende und neue Implementierung auf Konsistenz geprüft und die vorgeschlagenen algorithmischen Änderungen im Hinblick auf Veränderungen beim Erzeugen von Modellen und der Abstraktionsfähigkeit bei der Exploration unbekannter Daten untersucht.
Aus den Ergebnissen werden Empfehlungen zu konkreten Einstellungen sowie Vorschläge für weitere Untersuchungen abgeleitet. Die neue Implementierung von conML in Julia bietet im Hinblick auf Performanz, Flexibilität und Erweiterbarkeit einen guten Ausgangspunkt für weitere Forschung und Anwendung des Systems.:Contents
Abstract . . . . . III
Zusammenfassung . . . . . IV
Danksagung . . . . . V
Selbstständigkeitserklärung . . . . . V
1. Introduction
1.1. Research Questions . . . . . 2
2. Related Work
2.1. Hybrid AI Systems . . . . . 5
2.2. Constructivist Machine Learning (conML) . . . . . 6
2.3. Implemented Methods . . . . . 9
2.3.1. Unsupervised Machine Learning . . . . . 9
2.3.2. Supervised Machine Learning . . . . . 11
2.3.3. Supervised Feature Selection . . . . . 13
2.3.4. Unsupervised Feature Selection . . . . . 17
3. Methods and Implementation
3.1. Notable Algorithmic Changes . . . . . 19
3.1.1. Rescaling of Target Values . . . . . 19
3.1.2. ExtendedWinner Selection . . . . . 21
3.2. Package Structure . . . . . 23
3.3. Interfaces and Implementation of Specific Methods . . . . . 29
3.4. Datasets . . . . . 41
4. Results
4.1. Validation Against the conML Prototype . . . . . 43
4.2. Change in Abstraction Capability . . . . . 49
4.2.1. Influence of Target Scaling . . . . . 49
4.2.2. Influence of the Parameter kappa_p . . . . . 55
4.2.3. Influence of the Winner Selection Procedure . . . . . 61
5. Discussion
5.1. Reproduction Results . . . . . 67
5.2. Rescaling of Constructed Targets . . . . . 69
5.3. kappa_p and the Selection of Winner Models . . . . . 71
6. Conclusions
6.1. Contributions of this Work . . . . . 77
6.2. Future Work . . . . . 78
A. Julia Language Reference . . . . . 81
B. Additional Code Listings . . . . . 91
C. Available Parameters . . . . . 99
C.1. Block Processing . . . . . 105
D. Configurations Reference . . . . . 107
D.1. Unsupervised Methods . . . . . 107
D.2. Supervised Methods . . . . . 108
D.3. Feature Selection . . . . . 109
D.4. Winner Selection . . . . . 110
D.5. General Settings . . . . . 110
E. Supplemental Figures . . . . . 113
E.1. Replacing MAPE with RMSE for Z-Transform Target Scaling . . . . . 113
E.2. Combining Target Rescaling, Winner Selection and High kappa_p . . . . . 119
Bibliography . . . . . 123
List of Figures . . . . . 129
List of Listings . . . . . 133
List of Tables . . . . . 135
|
104 |
Behavior-specific proprioception models for robotic force estimation: a machine learning approachBerger, Erik 21 August 2018 (has links)
Robots that support humans in physically demanding tasks require accurate force sensing capabilities. A common way to achieve this is by monitoring the interaction with the environment directly with dedicated force sensors. Major drawbacks of such special purpose sensors are the increased costs and the reduced payload of the robot platform. Instead, this thesis investigates how the functionality of such sensors can be approximated by utilizing force estimation approaches. Most of today’s robots are equipped with rich proprioceptive sensing capabilities where even a robotic arm, e.g., the UR5, provides access to more than hundred sensor readings. Following this trend, it is getting feasible to utilize a wide variety of sensors for force estimation purposes. Human proprioception allows estimating forces such as the weight of an object by prior experience about sensory-motor patterns. Applying a similar approach to robots enables them to learn from previous demonstrations without the need of dedicated force sensors.
This thesis introduces Behavior-Specific Proprioception Models (BSPMs), a novel concept for enhancing robotic behavior with estimates of the expected proprioceptive feedback. A main methodological contribution is the operationalization of the BSPM approach using data-driven machine learning techniques. During a training phase, the behavior is continuously executed while recording proprioceptive sensor readings. The training data acquired from these demonstrations represents ground truth about behavior-specific sensory-motor experiences, i.e., the influence of performed actions and environmental conditions on the proprioceptive feedback. This data acquisition procedure does not require expert knowledge about the particular robot platform, e.g., kinematic chains or mass distribution, which is a major advantage over analytical approaches. The training data is then used to learn BSPMs, e.g. using lazy learning techniques or artificial neural networks. At runtime, the BSPMs provide estimates of the proprioceptive feedback that can be compared to actual sensations. The BSPM approach thus extends classical programming by demonstrations methods where only movement data is learned and enables robots to accurately estimate forces during behavior execution.
|
105 |
Dreidimensionales optisches Messsystem für eine effizientere geometrische Maschinenbeurteilung /Schmalzried, Siegfried. January 2007 (has links)
Zugl.: Karlsruhe, Universiẗat, Diss., 2007.
|
106 |
Anwendung von Verfahren des maschinellen Lernens und von evolutionären Algorithmen bei der räumlichen InferenzGips, Carsten January 2007 (has links)
Zugl.: Berlin, Techn. Univ., Diss., 2007
|
107 |
Analyzing biological expression data based on decision tree inductionFlöter, André January 2005 (has links)
<P>Modern biological analysis techniques supply scientists with various forms of data. One category of such data are the so called "expression data". These data indicate the quantities of biochemical compounds present in tissue samples.</P>
<P>Recently, expression data can be generated at a high speed. This leads in turn to amounts of data no longer analysable by classical statistical techniques. Systems biology is the new field that focuses on the modelling of this information.</P>
<P>At present, various methods are used for this purpose. One superordinate class of these methods is machine learning. Methods of this kind had, until recently, predominantly been used for classification and prediction tasks. This neglected a powerful secondary benefit: the ability to induce interpretable models.</P>
<P>Obtaining such models from data has become a key issue within Systems biology. Numerous approaches have been proposed and intensively discussed. This thesis focuses on the examination and exploitation of one basic technique: decision trees.</P>
<P>The concept of comparing sets of decision trees is developed. This method offers the possibility of identifying significant thresholds in continuous or discrete valued attributes through their corresponding set of decision trees. Finding significant thresholds in attributes is a means of identifying states in living organisms. Knowing about states is an invaluable clue to the understanding of dynamic processes in organisms. Applied to metabolite concentration data, the proposed method was able to identify states which were not found with conventional techniques for threshold extraction.</P>
<P>A second approach exploits the structure of sets of decision trees for the discovery of combinatorial dependencies between attributes. Previous work on this issue has focused either on expensive computational methods or the interpretation of single decision trees a very limited exploitation of the data. This has led to incomplete or unstable results. That is why a new method is developed that uses sets of decision trees to overcome these limitations.</P>
<P>Both the introduced methods are available as software tools. They can be applied consecutively or separately. That way they make up a package of analytical tools that usefully supplement existing methods.</P>
<P>By means of these tools, the newly introduced methods were able to confirm existing knowledge and to suggest interesting and new relationships between metabolites.</P> / <P>Neuere biologische Analysetechniken liefern Forschern verschiedenste Arten von Daten. Eine Art dieser Daten sind die so genannten "Expressionsdaten". Sie geben die Konzentrationen biochemischer Inhaltsstoffe in Gewebeproben an.<P>
<P>Neuerdings können Expressionsdaten sehr schnell erzeugt werden. Das führt wiederum zu so großen Datenmengen, dass sie nicht mehr mit klassischen statistischen Verfahren analysiert werden können. "System biology" ist eine neue Disziplin, die sich mit der Modellierung solcher Information befasst.</P>
<P>Zur Zeit werden dazu verschiedenste Methoden benutzt. Eine Superklasse dieser Methoden ist das maschinelle Lernen. Dieses wurde bis vor kurzem ausschließlich zum Klassifizieren und zum Vorhersagen genutzt. Dabei wurde eine wichtige zweite Eigenschaft vernachlässigt, nämlich die Möglichkeit zum Erlernen von interpretierbaren Modellen.</P>
<P>Die Erstellung solcher Modelle hat mittlerweile eine Schlüsselrolle in der "Systems biology" erlangt. Es sind bereits zahlreiche Methoden dazu vorgeschlagen und diskutiert worden. Die vorliegende Arbeit befasst sich mit der Untersuchung und Nutzung einer ganz grundlegenden Technik: den Entscheidungsbäumen.</P>
<P>Zunächst wird ein Konzept zum Vergleich von Baummengen entwickelt, welches das Erkennen bedeutsamer Schwellwerte in reellwertigen Daten anhand ihrer zugehörigen Entscheidungswälder ermöglicht. Das Erkennen solcher Schwellwerte dient dem Verständnis von dynamischen Abläufen in lebenden Organismen. Bei der Anwendung dieser Technik auf metabolische Konzentrationsdaten wurden bereits Zustände erkannt, die nicht mit herkömmlichen Techniken entdeckt werden konnten.</P>
<P>Ein zweiter Ansatz befasst sich mit der Auswertung der Struktur von Entscheidungswäldern zur Entdeckung von kombinatorischen Abhängigkeiten zwischen Attributen. Bisherige Arbeiten hierzu befassten sich vornehmlich mit rechenintensiven Verfahren oder mit einzelnen Entscheidungsbäumen, eine sehr eingeschränkte Ausbeutung der Daten. Das führte dann entweder zu unvollständigen oder instabilen Ergebnissen. Darum wird hier eine Methode entwickelt, die Mengen
von Entscheidungsbäumen nutzt, um diese Beschränkungen zu überwinden.</P>
<P>Beide vorgestellten Verfahren gibt es als Werkzeuge für den Computer, die entweder hintereinander oder einzeln verwendet werden können. Auf diese Weise stellen sie eine sinnvolle Ergänzung zu vorhandenen Analyswerkzeugen dar.</P>
<P>Mit Hilfe der bereitgestellten Software war es möglich, bekanntes Wissen zu bestätigen und interessante neue Zusammenhänge im Stoffwechsel von Pflanzen aufzuzeigen.</P>
|
108 |
Applications and extensions of Random Forests in genetic and environmental studiesMichaelson, Jacob 10 January 2011 (has links) (PDF)
Transcriptional regulation refers to the molecular systems that control the concentration of mRNA species within the cell. Variation in these controlling systems is not only responsible for many diseases, but also contributes to the vast phenotypic diversity in the biological world. There are powerful experimental approaches to probe these regulatory systems, and the focus of my doctoral research has been to develop and apply effective computational methods that exploit these rich data sets more completely. First, I present a method for mapping genetic regulators of gene expression (expression quantitative trait loci, or eQTL) using Random Forests. This approach allows for flexible modeling and feature selection, and results in eQTL that are more biologically supportable than those mapped with competing methods. Next, I present a method that finds interactions between genes that in turn regulate the expression of other genes. This is accomplished by finding recurring decision motifs in the forest structure that represent dependencies between genetic loci. Third, I present a method to use distributional differences in eQTL data to establish the regulatory roles of genes relative to other disease-associated genes. Using this method, we found that genes that are master regulators of other disease genes are more likely to be consistently associated with the disease in genetic association studies. Finally, I present a novel application of Random Forests to determine the mode of regulation of toxin-perturbed genes, using time-resolved gene expression. The results demonstrate a novel approach to supervised weighted clustering of gene expression data.
|
109 |
Modelle zur multisensoriellen Erfassung des Fahrzeugumfeldes mit Hilfe von Schätzverfahren /Cramer, Heiko. January 2006 (has links) (PDF)
Techn. Universiẗat, Diss.--Chemnitz, 2005.
|
110 |
Sprachsynthese und Spracherkennung mit gemeinsamen Datenbasen : akustische Analyse und Modellierung /Eichner, Matthias. January 2007 (has links) (PDF)
Techn. Universiẗat, Diss.--Dresden, 2006.
|
Page generated in 0.0639 seconds