335598 research outputs found
Sort by
Incorporating Causal Prior Knowledge into Deep Neural Networks
No full textDeep Neural Networks have achieved significant success in solving complex problems across various domains due to their ability to capture complicated patterns in large datasets; however, they often require large amounts of data to learn effectively and often lack transparency in their decision-making processes, relying heavily on correlation rather than causation. Such limitations have led
to incorporating causal Prior Knowledge into neural network models which stands as a significant advancement in machine learning, such knowledge can mitigate this data dependency, guide the learning process, and enhance not only the robustness and generalizability of models but also their interpretability and explainability. Additionally, it enables models to adapt to new tasks and domains with greater ease and effectiveness.
This report tackles the importance of incorporating causal prior knowledge into deep neural networks and the methodologies that facilitate this incorporation. Fundamental concepts of causality are reviewed, with emphasis on its importance for advancing AI towards causal representation learning
Blognotiz 14.12.2011 – Erinnerung an die fünf Platanen vom Schramberger Rathausplatz
No full textDieser Beitrag erforscht die kulturelle und symbolische Bedeutung der fünf Platanen (Platanus spp.), die einst den Rathausplatz in Schramberg säumten und im Dezember 2011 im Rahmen des städtebaulichen Projekts „Schrambergs Neue Mitte“ gefällt wurden. Über ihre schattenspendende Funktion hinaus verkörperten diese Bäume einen Hauch mediterraner Atmosphäre in der sonst eher kühl-feuchten Landschaft des Schwarzwalds und standen symbolisch für die kollektive Erinnerung und Identität der lokalen Gemeinschaft. Anhand persönlicher Erinnerungen, regionaler Medienberichte und literarischer Bezüge – insbesondere François Mitterrands tiefe Verbindung zu Bäumen und Landschaften, wie sie in Robert Schneiders „Les Mitterrand“ beschrieben wird – werden die Platanen als integrale Bestandteile der „paysages géographiques“ und Kulturlandschaften interpretiert. Ihr Verlust wird als Metapher für den Wandel urbaner Räume, das kulturelle Erbe und die Erosion historischer Identität zugunsten der Modernisierung analysiert. Der Artikel lenkt den Blick zudem auf andere mediterrane botanische Relikte in der Region Schramberg, wie die Edelkastanie (Castanea sativa) und den Judasbaum (Cercis siliquastrum), die als letzte Zeugnisse einer vergangenen Epoche fortbestehen. Abschließend plädiert der Beitrag für eine bewusste Einbindung solcher Elemente in die zeitgenössische Stadtplanung, um kulturelle Kontinuität und regionale Eigenart zu bewahren
Adding 161Dy-Mössbauer Spectroscopy to a Multitechnique Investigation of Magnetic Transitions in a {CoIII3DyIII3} Single-Molecule Toroic
No full textMulti-Flow Process Mining as an Enabler for Comprehensive Digital Twins of Manufacturing Systems
No full textProcess Mining (PM) has proven useful for extracting Digital Twin (DT) simulation models for manufacturing systems. PM is a family of approaches designed to capture temporal process flows by analyzing event logs that contain time-stamped records of relevant events. With the widespread availability of sensors in modern manufacturing systems, events can be tracked across multiple process dimensions beyond time, enabling a more comprehensive performance analysis. Some of these dimensions include energy and waste. By integrating and treating these dimensions analogously to time, we enable the use of PM to extract process flows along multiple dimensions, an approach we refer to as multi-flow PM. The resulting models that capture multiple dimensions are ultimately combined to enable comprehensive DTs that support multi-objective decision-making. In this paper, we present our approach to generating these multidimensional discrete-event models and, through an illustrative case study, demonstrate how they can be utilized for multi-objective decision support
Formal Process Maturity Measure
No full textMeasuring the maturity of a process instance is essential because we can evaluate the progress toward a mature production process; Consider we have added some sensors or actuators in the manufacturing process so we need a formal basement for measuring the maturity of the process after these changes. We used the definitions in the control theory to provide a formal measure for process maturity. We defined Elucidability E, Forcability F , and Supervisability S that are ostensive, interpretable, and based on quantities that can be determined or estimated. These lead to a formal definition for a measure of the process maturity M , that combines technical and economic considerations
Improved Copper‐Zinc Based Catalysts for the Partial Dehydrogenation of Dicyclohexylmethanol
No full textCu/ZnO/ZrO 2 (CZZ) catalysts outperform conventional Cu/ZnO/Al 2 O3 (CZA) materials in methanol synthesis. Building on recent findings that CZA catalysts enable hydrogen release from the oxygen-containing LOHC compound dicyclohexylmethanol (H14BP) below 200◦C, this study investigates structure-activity correlations with CZZ catalysts in the partial dehydrogenation of H14-BP as a model reaction. CZZ materials were produced by continuous co-precipitation with subsequent batch suspension ageing. A ZrO2 content between 4 and 8 mol% increased the specific surface area and catalytic activity. Zn-rich materials with elevated aurichalcite [(Cu,Zn) 5 (OH) 6 (CO3 ) 2 ] content in the catalyst precursor achieved higher activity despite a reduced specific surface area. Ageing at 70◦C promoted aurichalcite formation and improved performance, whereas higher temperatures reduced the specific surface area. An initial pH value of 6.7 enhanced Zn uptake during ageing and increased dehydrogenation productivity by 45% compared to pH 7.1. High catalyst productivity correlated with aurichalcite contents up to 98% and small crystallite sizes. Overall, CZZ outperformed CZA catalysts in the partial dehydrogenation of H14-BP, with the aurichalcite phase playing a crucial role. Our results demonstrate the potential of this material class for selective dehydrogenation reactions and enable targeted further development based on the correlation between material-specific properties and catalytic activit
In-reservoir monitoring and modelling concepts to assess the sedimentation dynamics of Enguri reservoir
No full textIn this study we combined monitoring and modelling techniques to understand sediment dynamics and improve management strategies in the Enguri reservoir. Methods included topographic differencing, sub-bottom profiling, water sampling, and visual inspection. A Delft3D model simulated sediment transport under various conditions. Topographic differencing indicated a 16% loss in reservoir volume, primarily in the dead storage zone, mainly due to the operation regime. Sub-bottom profiling effectively quantified silt and clay deposition (43% of the total sediment), while showing difficulties in penetrating high density sand and gravel layers. In total 29 modelling scenarios were investigated. Modelling identified reservoir head as a key factor in sedimentation patterns. Partial flushing was found suboptimal, leading to movement of large volumes of sediment within the reservoir. In this case, empty flushing is recommended as the most effective mitigation measure. The combination of monitoring with modelling provided a framework for understanding and managing sediment issues in large reservoirs such as Enguri
Block copolymer concepts of how transcription organizes the stem cell genome
No full textStem cells display a highly dispersed genome organization that supports flexible gene regulation. Here, we present block copolymer concepts to explore how transcriptional activity from specific genomic regions, or ‘blocks’, shapes and controls several features of this architecture. Nascent transcripts tethered to chromatin can disrupt compaction and promote the formation of a micro-dispersed state of euchromatin, explaining one typical feature of the stem cell genome. A second feature is long-lived transcriptional clusters, which form via condensation at super-enhancer blocks and mediate both long-range interactions and local transcription factor accumulation. Lastly, we conceptualize promoters and gene bodies as a two-block polymer, for which sequential switching on and off of the polymer blocks controls the association and subsequent release of developmental genes with the long-lived clusters. The presented block copolymer framework provides explanations as well as hypotheses of how transcription-associated processes contribute to distinct features of stem cell genome organization
The effect of aging precipitation on the fretting wear behavior of 17-4 PH stainless steel used for tight fit assemblies
No full textAn experimental study on fretting wear behavior of a 17-4 PH stainless steel aging heat treated to different microstructural features with different precipitate variants is presented. The fretting wear tests with respect to different fretting running regimes were carried out. The microstructural observations show that below 460 °C of aging temperature, only NbC precipitates were observed. While as the aging temperature reached the 460 °C, not only NbC but also Cu-rich precipitates (CRPs) can be found, and even more so at higher aging temperature. Wear results indicate that in the partial slip regime (PSR) and mixed fretting regime (MFR), the fretting wear volume decreases with the aging temperature increases up to 410 °C, beyond which the wear volume commences to arise. The sample T2 (410 °C) has a highest fretting wear resistance. In contrast, in gross slip regime (GSR), although the correlation of fretting wear volume and aging temperature shows a similar trend with that in PSR and MFR, the sample T3 (460 °C) has a highest fretting wear resistance in GSR. It suggests that the precipitates have significant effect on fretting wear resistance, and their effects are fretting running regime dependent. Subsurface observations demonstrate that fine-size CRPs could promote the formation of the compacted and oxide-contained plastic deformation layer, which can improve the fretting wear resistance. The results can provide a good guideline to optimize the target microstructure for a specific working condition or tune the working condition parameter to allow a best fretting wear resistance