727 research outputs found
Sort by
Parametrizing Analog Multi-Compartment Neurons with Genetic Algorithms [Data]
No full textThis data is presented in the paper: "Parametrizing Analog Multi-Compartment Neurons with Genetic Algorithms" which is currently under review.
Further information about the contents of the files can be found in the `README.md`.
Abstract:
This paper employs genetic algorithms to parameterize the leak conductance and inter-compartment conductance of multi-compartment neurons on the analog BrainScaleS-2 neuromorphic hardware platform. These parameters are not always directly derivable from neuron observations but are crucial for replicating desired observations. Genetic algorithms promise parameterization without domain knowledge of the neuromorphic substrate or underlying neuron model. The objective of this study is to replicate the attenuation behavior of an excitatory postsynaptic potential (EPSP) traveling along a linear chain of compartments, which was observed to exhibit an exponential decay of the EPSP’s amplitude. A comprehensive grid search was conducted to evaluate the solutions from the genetic algorithm. To counteract trial-to-trial variations in analog systems, spike-triggered averaging was utilized. The study demonstrated the multi-objective search capabilities of genetic algorithms, allowing for the constraint of multiple parameters to reach multiple target observables. The algorithm successfully replicated the desired EPSP attenuation behavior in both single and multi-objective searches illustrating the applicability of genetic algorithms to parameterize analog neuromorphic hardware
Bright Circularly-Polarized Photoluminescence in Chiral Layered Lead-Halide Perovskites [Research Data]
No full textHybrid semiconductor materials are predicted to lock chirality into place and encode asymmetry into their electronic states, while the softness of their crystal lattice accommodates lattice strain and maintains high crystal quality with the low defect densities necessary for high luminescence yields. The realization of chiral bulk emitters with bright circularly-polarized luminescence from such materials is desired for the design of chiroptical photonic and opto-spintronic applications. Here, we report photoluminescence quantum efficiencies (PLQE) as high as 39%, and degrees of circularly polarized photoluminescence of up to 52%, at room temperature, in the chiral layered hybrid lead-halide perovskites (R/S/Rac)-3BrMBA2PbI4 (3BrMBA = 1-(3-Bromphenyl)-ethylamine). Using x-ray diffraction and density-functional theory we elucidate the detailed chirality transfer mechanism from chiral crystal structures to spin-orbit-split band structures. Using state-of-the-art transient chiroptical spectroscopy, we rationalize the excellent photoluminescence yields from suppression of non-radiative loss channels and very high rates of radiative recombination. We further find that photo-excitations sustain polarization lifetimes that exceed the timescales of radiative decays, which rationalize the high degrees of polarized luminescence. We postulate that the superior optoelectronic properties of the layered hybrid perovskites arise from their special tolerance to crystal structure chirality, which we carefully designed by cation engineering. Our findings pave the way towards high-performance solution-processed photonic systems for chiroptical applications and chiral-spintronic logic at room temperature
Ergänzungsmaterial zu: Settlement patterns of the Middle Palaeolithic in Southern Germany. A GIS-supported predictive model for sites in Bavaria and Baden-Wurttemberg / Das Siedlungsmuster des Mittelpaläolithikums in Süddeutschland. Eine GIS-gestützte Archäoprognose für Fundstellen in Bayern und Baden-Württemberg
No full textAs a largely glacier-free zone throughout the entire Middle Palaeolithic, Southern Germany has often been discussed as a key area for Neanderthal migration. In this study, the settlement patterns of the Southern German Middle Palaeolithic were investigated via Weighted Layer Analysis, resulting in key insights about the prognostic qualities of topographic variables like elevation, slope, aspect, distance to river and outgoing visibility, as well as two predictive maps for cave and open-air sites. Comparing the high probability zones for both site types, their possible interplay in Southern Germany and the special role of the infrastructure of the Franconian Swabian Jura for Neanderthal migration in Europe are discussed
Optimising Urban Measurement Networks for CO2 Flux Estimation: A High-Resolution Observing System Simulation Experiment using GRAMM/GRAL [data]
No full textThe data set contains CO2 concentration fields at 2m height above ground in Heidelberg, Germany as simulated by the model GRAMM (v19.1)/GRAL(v.19.1).
It can be used as input for Observing System Simulation Experiments (OSSE).
The model GRAMM and a link to its documentation can be found here: https://github.com/GralDispersionModel/GRAMM
The model GRAL and a link to the documentation can be found here: https://github.com/GralDispersionModel/GRAL
GRAL uses the GRAMM wind fields as mesoscale input and refines the wind fields to a higher resolution taking into account the flow around buildings. The GRAL domain size is 12.3km x12.3km with 10m x 10m horizontal resolution and 2m vertical resolution. A Lagrangian particle simulation is performed to obtain the hourly steady state particle distribution from emissions, which is used to compute the concentration field
Dimerization of a Reactive Azaacene Diradical: Synthesis of a Covalent Azaacene Cage [Data]
No full textTwo series of regioisomeric dicyanomethylene substituted dithienodiazatetracenes with formal para- or ortho- quinodimethane subunits were synthesized and characterized. Whereas the para-isomers (p-n, diradical index y0=0.01) are stable and isolable, the ortho-isomer (y0=0.98) dimerizes into a covalent azaacene cage. Four elongated σ-CC bonds are formed and the former triisopropylsilyl(TIPS)-ethynylene groups transformed into cumulene-units. The azaacene cage dimer (o-1)2 was characterized by X-ray single crystal structure analysis and temperature-dependent infrared (IR), electron paramagnetic resonance (EPR, solid state), nuclear magnetic resonance (NMR) and ultraviolet-visible (UV-Vis) spectroscopies (solution) indicating reformation of o-1
Beteiligung von Patientinnen und Patienten an Klinischen Studien
No full textIn diesem Dokument wird anhand von Beispielen und Methoden dargestellt, wie die aktive Beteiligung von Patientinnen und Patienten an Klinischen Studien in unterschiedlichen Projektphasen (Konzeption, Durchführung und Auswertung) geplant und unterstützt werden kann
prometheus Source Code
No full textThe prometheus digital image archive software is the driving force behind the
prometheus image archive. It's being developed since late fall of 2006, on top of the Ruby on Rails framework. It is Free Software, released under the terms of the GNU Affero General Public License
Substituting density functional theory in reaction barrier calculations for hydrogen atom transfer in proteins [data]
No full textHydrogen atom transfer (HAT) reactions are important in many biological systems. As these reactions are hard to observe experimentally, it is of high interest to shed light on them using simulations. Here, we present a machine learning model based on graph neural networks for the prediction of energy barriers of HAT reactions in proteins. As input, the model uses exclusively non-optimized structures as obtained from classical simulations. It was trained on more than 17,000 energy barriers calculated using hybrid density functional theory. We built and evaluated the model in the context of HAT in collagen, but we show that the same workflow can easily be applied to HAT reactions in other biological or synthetic polymers. We obtain for relevant reactions (small reaction distances) a model with good predictive power (R2 ∼ 0.9 and mean absolute error of < 3 kcal/mol). As the inference speed is high, this model enables evaluations of dozens of chemical situations within seconds. When combined with molecular dynamics in a kinetic Monte-Carlo scheme, the model paves the way toward reactive simulations
[Ergänzungsmaterial zu] Rinne, Christoph: Odagsen und Großenrode, Ldkr. Northeim. Jungsteinzeitliche Kollektivgräber im südlichen Leinetal, Rahden/Westf. 2003
No full textSpatialite Database (SQL) and CSV files with data of the Neolithic collective grave Odagsen I in Lower Saxony (Germany). Spatial data from individual hand-drawn plan sheets of the excavation from 1980 to 1984 digitised in AutoCAD 1996. The data were revised in 2019 and transferred via the dxf files to a spatialite database. The find lists of the excavation were digitised into an Excel spreadsheet in 1996. This table was transformed with R package reshape2::melt, data rows with column name and value are now available for each find number
Classification of structural building damage grades from multi-temporal photogrammetric point clouds using a machine learning model trained on virtual laser scanning data [Data and Source Code]
No full textAutomatic damage assessment by analysing UAV-derived 3D point clouds provides fast information on the damage situation after an earthquake. However, the assessment of different damage grades is challenging given the variety in damage characteristics and limited transferability of methods to other geographic regions or data sources. We present a novel change-based approach to automatically assess multi-class building damage from real-world point clouds using a machine learning model trained on virtual laser scanning (VLS) data. Therein, we (1) identify object-specific point cloud-based change features, (2) extract changed building parts using k-means clustering, (3) train a random forest machine learning model with VLS data based on object-specific change features, and (4) use the classifier to assess building damage in real-world photogrammetric point clouds. We evaluate the classifier with respect to its capacity to classify three damage grades (heavy, extreme, destruction) in pre-event and post-event point clouds of an earthquake in L’Aquila (Italy). Using object-specific change features derived from bi-temporal point clouds, our approach is transferable with respect to multi-source input point clouds used for model training (VLS) and application (real-world photogrammetry). We further achieve geographic transferability by using simulated training data which characterises damage grades across different geographic regions. The model yields high multi-target classification accuracies (overall accuracy: 92.0%–95.1%). Classification performance improves only slightly when using real-world region-specific training data (3% higher overall accuracies). We consider our approach especially relevant for applications where timely information on the damage situation is required and sufficient real-world training data is not available.
This dataset includes
3D building models (building_models.zip) representing the target damage grades (no damage, heavy damage, extreme damage, destruction) of this study
Python source code (code.zip) used in this study to (1) generate simulated multi-temporal 3D point clouds using HELIOS++ (https://github.com/3dgeo-heidelberg/helios), (2) extract damaged building parts using k-means clustering, (3) compute object-specific geometric change features per building (4) train a multi-target random forest classifier to classify buildings into four damage grades based on object-specific change features.
</ul