Refine
Departments, institutes and facilities
- Fachbereich Informatik (50)
- Fachbereich Ingenieurwissenschaften und Kommunikation (22)
- Institut für Technik, Ressourcenschonung und Energieeffizienz (TREE) (17)
- Fachbereich Angewandte Naturwissenschaften (9)
- Fachbereich Wirtschaftswissenschaften (6)
- Institut für Cyber Security & Privacy (ICSP) (4)
- Internationales Zentrum für Nachhaltige Entwicklung (IZNE) (2)
- Fachbereich Sozialpolitik und Soziale Sicherung (1)
- Institut für Medienentwicklung und -analyse (IMEA) (1)
- Institut für funktionale Gen-Analytik (IFGA) (1)
Document Type
- Preprint (90) (remove)
Year of publication
Keywords
- Evolutionary Computation (2)
- FOS: Computer and information sciences (2)
- burnout (2)
- inborn error of metabolism (2)
- ketone body (2)
- lignin (2)
- metabolic acidosis (2)
- metabolic decompensation (2)
- organic aciduria (2)
- psychological detachment (2)
Renewable resources gain increasing interest as source for environmentally benign biomaterials, such as drug encapsulation/release compounds, and scaffolds for tissue engineering in regenerative medicine. Being the second largest naturally abundant polymer, the interest in lignin valorization for biomedical utilization is rapidly growing. Depending on resource and isolation procedure, lignin shows specific antioxidant and antimicrobial activity. Today, efforts in research and industry are directed toward lignin utilization as renewable macromolecular building block for the preparation of polymeric drug encapsulation and scaffold materials. Within the last five years, remarkable progress has been made in isolation, functionalization and modification of lignin and lignin-derived compounds. However, literature so far mainly focuses lignin-derived fuels, lubricants and resins. The purpose of this review is to summarize the current state of the art and to highlight the most important results in the field of lignin-based materials for potential use in biomedicine (reported in 2014–2018). Special focus is drawn on lignin-derived nanomaterials for drug encapsulation and release as well as lignin hybrid materials used as scaffolds for guided bone regeneration in stem cell-based therapies.
Antioxidant activity is an essential feature required for oxygen-sensitive merchandise and goods, such as food and corresponding packaging as well as materials used in cosmetics and biomedicine. For example, vanillin, one of the most prominent antioxidants, is fabricated from lignin, the second most abundant natural polymer in the world. Antioxidant potential is primarily related to the termination of oxidation propagation reactions through hydrogen transfer. The application of technical lignin as a natural antioxidant has not yet been implemented in the industrial sector, mainly due to the complex heterogeneous structure and polydispersity of lignin. Thus, current research focuses on various isolation and purification strategies to improve the compatibility of lignin material with substrates and enhancing its stabilizing effect.
Current robot platforms are being employed to collaborate with humans in a wide range of domestic and industrial tasks. These environments require autonomous systems that are able to classify and communicate anomalous situations such as fires, injured persons, car accidents; or generally, any potentially dangerous situation for humans. In this paper we introduce an anomaly detection dataset for the purpose of robot applications as well as the design and implementation of a deep learning architecture that classifies and describes dangerous situations using only a single image as input. We report a classification accuracy of 97 % and METEOR score of 16.2. We will make the dataset publicly available after this paper is accepted.
In this paper we introduce the Perception for Autonomous Systems (PAZ) software library. PAZ is a hierarchical perception library that allow users to manipulate multiple levels of abstraction in accordance to their requirements or skill level. More specifically, PAZ is divided into three hierarchical levels which we refer to as pipelines, processors, and backends. These abstractions allows users to compose functions in a hierarchical modular scheme that can be applied for preprocessing, data-augmentation, prediction and postprocessing of inputs and outputs of machine learning (ML) models. PAZ uses these abstractions to build reusable training and prediction pipelines for multiple robot perception tasks such as: 2D keypoint estimation, 2D object detection, 3D keypoint discovery, 6D pose estimation, emotion classification, face recognition, instance segmentation, and attention mechanisms.
In this paper we propose an implement a general convolutional neural network (CNN) building framework for designing real-time CNNs. We validate our models by creating a real-time vision system which accomplishes the tasks of face detection, gender classification and emotion classification simultaneously in one blended step using our proposed CNN architecture. After presenting the details of the training procedure setup we proceed to evaluate on standard benchmark sets. We report accuracies of 96% in the IMDB gender dataset and 66% in the FER-2013 emotion dataset. Along with this we also introduced the very recent real-time enabled guided back-propagation visualization technique. Guided back-propagation uncovers the dynamics of the weight changes and evaluates the learned features. We argue that the careful implementation of modern CNN architectures, the use of the current regularization methods and the visualization of previously hidden features are necessary in order to reduce the gap between slow performances and real-time architectures. Our system has been validated by its deployment on a Care-O-bot 3 robot used during RoboCup@Home competitions. All our code, demos and pre-trained architectures have been released under an open-source license in our public repository.
Humans exhibit flexible and robust behavior in achieving their goals. We make suitable substitutions for objects, actions, or tools to get the job done. When opportunities that would allow us to reach our goals with less effort arise, we often take advantage of them. Robots are not nearly as robust in handling such situations. Enabling a domestic service robot to find ways to get a job done by making substitutions is the goal of our work. In this paper, we highlight the challenges faced in our approach to combine Hierarchical Task Network planning, Description Logics, and the notions of affordances and conceptual similarity. We present open questions in modeling the necessary knowledge, creating planning problems, and enabling the system to handle cases where plan generation fails due to missing/unavailable objects.
Representing 3D surfaces as level sets of continuous functions over R3 is the common denominator of neural implicit representations, which recently enabled remarkable progress in geometric deep learning and computer vision tasks. In order to represent 3D motion within this framework, it is often assumed (either explicitly or implicitly) that the transformations which a surface may undergo are homeomorphic: this is not necessarily true, for instance, in the case of fluid dynamics. In order to represent more general classes of deformations, we propose to apply this theoretical framework as regularizers for the optimization of simple 4D implicit functions (such as signed distance fields). We show that our representation is capable of capturing both homeomorphic and topology-changing deformations, while also defining correspondences over the continuously-reconstructed surfaces.
The majority of biomedical knowledge is stored in structured databases or as unstructured text in scientific publications. This vast amount of information has led to numerous machine learning-based biological applications using either text through natural language processing (NLP) or structured data through knowledge graph embedding models (KGEMs). However, representations based on a single modality are inherently limited. To generate better representations of biological knowledge, we propose STonKGs, a Sophisticated Transformer trained on biomedical text and Knowledge Graphs. This multimodal Transformer uses combined input sequences of structured information from KGs and unstructured text data from biomedical literature to learn joint representations. First, we pre-trained STonKGs on a knowledge base assembled by the Integrated Network and Dynamical Reasoning Assembler (INDRA) consisting of millions of text-triple pairs extracted from biomedical literature by multiple NLP systems. Then, we benchmarked STonKGs against two baseline models trained on either one of the modalities (i.e., text or KG) across eight different classification tasks, each corresponding to a different biological application. Our results demonstrate that STonKGs outperforms both baselines, especially on the more challenging tasks with respect to the number of classes, improving upon the F1-score of the best baseline by up to 0.083. Additionally, our pre-trained model as well as the model architecture can be adapted to various other transfer learning applications. Finally, the source code and pre-trained STonKGs models are available at https://github.com/stonkgs/stonkgs and https://huggingface.co/stonkgs/stonkgs-150k.
Solar photovoltaic power output is modulated by atmospheric aerosols and clouds and thus contains valuable information on the optical properties of the atmosphere. As a ground-based data source with high spatiotemporal resolution it has great potential to complement other ground-based solar irradiance measurements as well as those of weather models and satellites, thus leading to an improved characterisation of global horizontal irradiance. In this work several algorithms are presented that can retrieve global tilted and horizontal irradiance and atmospheric optical properties from solar photovoltaic data and/or pyranometer measurements. Specifically, the aerosol (cloud) optical depth is inferred during clear sky (completely overcast) conditions. The method is tested on data from two measurement campaigns that took place in Allgäu, Germany in autumn 2018 and summer 2019, and the results are compared with local pyranometer measurements as well as satellite and weather model data. Using power data measured at 1 Hz and averaged to 1 minute resolution, the hourly global horizontal irradiance is extracted with a mean bias error compared to concurrent pyranometer measurements of 11.45 W m−2, averaged over the two campaigns, whereas for the retrieval using coarser 15 minute power data the mean bias error is 16.39 W m−2.
During completely overcast periods the cloud optical depth is extracted from photovoltaic power using a lookup table method based on a one-dimensional radiative transfer simulation, and the results are compared to both satellite retrievals as well as data from the COSMO weather model. Potential applications of this approach for extracting cloud optical properties are discussed, as well as certain limitations, such as the representation of 3D radiative effects that occur under broken cloud conditions. In principle this method could provide an unprecedented amount of ground-based data on both irradiance and optical properties of the atmosphere, as long as the required photovoltaic power data are available and are properly pre-screened to remove unwanted artefacts in the signal. Possible solutions to this problem are discussed in the context of future work.
Suppose we have n keys, n access probabilities for the keys, and n+1 access probabilities for the gaps between the keys. Let h_min(n) be the minimal height of a binary search tree for n keys. We consider the problem to construct an optimal binary search tree with near minimal height, i.e.\ with height h <= h_min(n) + Delta for some fixed Delta. It is shown, that for any fixed Delta optimal binary search trees with near minimal height can be constructed in time O(n^2). This is as fast as in the unrestricted case. So far, the best known algorithms for the construction of height-restricted optimal binary search trees have running time O(L n^2), whereby L is the maximal permitted height. Compared to these algorithms our algorithm is at least faster by a factor of log n, because L is lower bounded by log n.
In this contribution, we perform computer simulations to expedite the development of hydrogen storages based on metal hydride. These simulations enable in-depth analysis of the processes within the systems which otherwise could not be achieved. That is, because the determination of crucial process properties require measurement instruments in the setup which are currently not available. Therefore, we investigate the reliability of reaction values that are determined by a design of experiments.
Specifically, we first explain our model setup in detail. We define the mathematical terms to obtain insights into the thermal processes and reaction kinetics. We then compare the simulated results to measurements of a 5-gram sample consisting of iron-titanium-manganese (FeTiMn) to obtain the values with the highest agreement with the experimental data. In addition, we improve the model by replacing the commonly used Van’t-Hoff equation by a mathematical expression of the pressure-composition-isotherms (PCI) to calculate the equilibrium pressure.
Finally, the parameters’ accuracy is checked in yet another with an existing metal hydride system. The simulated results demonstrate high concordance with experimental data, which advocate the usage of approximated kinetic reaction properties by a design of experiments for further design studies. Furthermore, we are able to determine process parameters like the entropy and enthalpy.
The lattice Boltzmann method (LBM) is an efficient simulation technique for computational fluid mechanics and beyond. It is based on a simple stream-and-collide algorithm on Cartesian grids, which is easily compatible with modern machine learning architectures. While it is becoming increasingly clear that deep learning can provide a decisive stimulus for classical simulation techniques, recent studies have not addressed possible connections between machine learning and LBM. Here, we introduce Lettuce, a PyTorch-based LBM code with a threefold aim. Lettuce enables GPU accelerated calculations with minimal source code, facilitates rapid prototyping of LBM models, and enables integrating LBM simulations with PyTorch's deep learning and automatic differentiation facility. As a proof of concept for combining machine learning with the LBM, a neural collision model is developed, trained on a doubly periodic shear layer and then transferred to a different flow, a decaying turbulence. We also exemplify the added benefit of PyTorch's automatic differentiation framework in flow control and optimization. To this end, the spectrum of a forced isotropic turbulence is maintained without further constraining the velocity field.
This paper addresses the classification of Arabic text data in the field of Natural Language Processing (NLP), with a particular focus on Natural Language Inference (NLI) and Contradiction Detection (CD). Arabic is considered a resource-poor language, meaning that there are few data sets available, which leads to limited availability of NLP methods. To overcome this limitation, we create a dedicated data set from publicly available resources. Subsequently, transformer-based machine learning models are being trained and evaluated. We find that a language-specific model (AraBERT) performs competitively with state-of-the-art multilingual approaches, when we apply linguistically informed pre-training methods such as Named Entity Recognition (NER). To our knowledge, this is the first large-scale evaluation for this task in Arabic, as well as the first application of multi-task pre-training in this context.
Fundamental hydrogen storage properties of TiFe-alloy with partial substitution of Fe by Ti and Mn
(2020)
TiFe intermetallic compound has been extensively studied, owing to its low cost, good volumetric hydrogen density, and easy tailoring of hydrogenation thermodynamics by elemental substitution. All these positive aspects make this material promising for large-scale applications of solid-state hydrogen storage. On the other hand, activation and kinetic issues should be amended and the role of elemental substitution should be further understood. This work investigates the thermodynamic changes induced by the variation of Ti content along the homogeneity range of the TiFe phase (Ti:Fe ratio from 1:1 to 1:0.9) and of the substitution of Mn for Fe between 0 and 5 at.%. In all considered alloys, the major phase is TiFe-type together with minor amounts of TiFe2 or \b{eta}-Ti-type and Ti4Fe2O-type at the Ti-poor and rich side of the TiFe phase domain, respectively. Thermodynamic data agree with the available literature but offer here a comprehensive picture of hydrogenation properties over an extended Ti and Mn compositional range. Moreover, it is demonstrated that Ti-rich alloys display enhanced storage capacities, as long as a limited amount of \b{eta}-Ti is formed. Both Mn and Ti substitutions increase the cell parameter by possibly substituting Fe, lowering the plateau pressures and decreasing the hysteresis of the isotherms. A full picture of the dependence of hydrogen storage properties as a function of the composition will be discussed, together with some observed correlations.
Today, more than 70 million tons of lignin are produced by the pulp and paper industry every year. However, the utilization of lignin as a source for chemical synthesis is still limited due to the complex and heterogeneous lignin structure. The purpose of this study was a selective photodegradation of industrially available kraft lignin in order to obtain appropriate fragments and building block chemicals for further utilization, e.g. polymerization. Thus, kraft lignin obtained from soft wood black liquor by acidification was dissolved in sodium hydroxide and irradiated at a wavelength of 254 nm with and without the presence of titanium dioxide in various concentrations. Analyses of the irradiated products via SEC showed decreasing molar masses and decreasing polydispersity indices over time. At the end of the irradiation period the lignin was depolymerised to form fragments as small as the lignin monomers. TOC analyses showed minimal mineralisation due to the depolymerisation process.
4GREAT is an extension of the German Receiver for Astronomy at Terahertz frequencies (GREAT) operated aboard the Stratospheric Observatory for Infrared Astronomy (SOFIA). The spectrometer comprises four different detector bands and their associated subsystems for simultaneous and fully independent science operation. All detector beams are co-aligned on the sky. The frequency bands of 4GREAT cover 491-635, 890-1090, 1240-1525 and 2490-2590 GHz, respectively. This paper presents the design and characterization of the instrument, and its in-flight performance. 4GREAT saw first light in June 2018, and has been offered to the interested SOFIA communities starting with observing cycle 6.
We derive rates of convergence for limit theorems that reveal the intricate structure of the phase transitions in a mean-field version of the Blume-Emery-Griffith model. The theorems consist of scaling limits for the total spin. The model depends on the inverse temperature β and the interaction strength K. The rates of convergence results are obtained as (β,K) converges along appropriate sequences (βn,Kn) to points belonging to various subsets of the phase diagram which include a curve of second-order points and a tricritical point. We apply Stein's method for normal and non-normal approximation avoiding the use of transforms and supplying bounds, such as those of Berry-Esseen quality, on approximation error. We observe an additional phase transition phenomenon in the sense that depending on how fast Kn and βn are converging to points in various subsets of the phase diagram, different rates of convergences to one and the same limiting distribution occur.
The latest trends in inverse rendering techniques for reconstruction use neural networks to learn 3D representations as neural fields. NeRF-based techniques fit multi-layer perceptrons (MLPs) to a set of training images to estimate a radiance field which can then be rendered from any virtual camera by means of volume rendering algorithms. Major drawbacks of these representations are the lack of well-defined surfaces and non-interactive rendering times, as wide and deep MLPs must be queried millions of times per single frame. These limitations have recently been singularly overcome, but managing to accomplish this simultaneously opens up new use cases. We present KiloNeuS, a new neural object representation that can be rendered in path-traced scenes at interactive frame rates. KiloNeuS enables the simulation of realistic light interactions between neural and classic primitives in shared scenes, and it demonstrably performs in real-time with plenty of room for future optimizations and extensions.
Recent experimental evidence suggest that mebendazole, a popular antiparasitic drug, binds to heat shock protein 90 (Hsp90) and inhibit acute myeloid leukemia cell growth. In this study we use quantum mechanics (QM), molecular similarity and molecular dynamics (MD) calculations to predict possible binding poses of mebendazole to the adenosine triphosphate (ATP) binding site of Hsp90. Extensive conformational searches and minimization of the five tautomers of mebendazole using MP2/aug-cc-pVTZ theory level resulting in 152 minima being identified. Mebendazole-Hsp90 complex models were created using the QM optimized conformations and protein coordinates obtained from experimental crystal structures that were chosen through similarity calculations. Nine different poses were identified from a total of 600 ns of explicit solvent, all-atom MD simulations using two different force fields. All simulations support the hypothesis that mebendazole is able to bind to the ATP binding site of Hsp90.
Comparative Evaluation of Pretrained Transfer Learning Models on Automatic Short Answer Grading
(2020)
Automatic Short Answer Grading (ASAG) is the process of grading the student answers by computational approaches given a question and the desired answer. Previous works implemented the methods of concept mapping, facet mapping, and some used the conventional word embeddings for extracting semantic features. They extracted multiple features manually to train on the corresponding datasets. We use pretrained embeddings of the transfer learning models, ELMo, BERT, GPT, and GPT-2 to assess their efficiency on this task. We train with a single feature, cosine similarity, extracted from the embeddings of these models. We compare the RMSE scores and correlation measurements of the four models with previous works on Mohler dataset. Our work demonstrates that ELMo outperformed the other three models. We also, briefly describe the four transfer learning models and conclude with the possible causes of poor results of transfer learning models.
The MAP-Elites algorithm produces a set of high-performing solutions that vary according to features defined by the user. This technique has the potential to be a powerful tool for design space exploration, but is limited by the need for numerous evaluations. The Surrogate-Assisted Illumination algorithm (SAIL), introduced here, integrates approximative models and intelligent sampling of the objective function to minimize the number of evaluations required by MAP-Elites.
The ability of SAIL to efficiently produce both accurate models and diverse high performing solutions is illustrated on a 2D airfoil design problem. The search space is divided into bins, each holding a design with a different combination of features. In each bin SAIL produces a better performing solution than MAP-Elites, and requires several orders of magnitude fewer evaluations. The CMA-ES algorithm was used to produce an optimal design in each bin: with the same number of evaluations required by CMA-ES to find a near-optimal solution in a single bin, SAIL finds solutions of similar quality in every bin.
The way solutions are represented, or encoded, is usually the result of domain knowledge and experience. In this work, we combine MAP-Elites with Variational Autoencoders to learn a Data-Driven Encoding (DDE) that captures the essence of the highest-performing solutions while still able to encode a wide array of solutions. Our approach learns this data-driven encoding during optimization by balancing between exploiting the DDE to generalize the knowledge contained in the current archive of elites and exploring new representations that are not yet captured by the DDE. Learning representation during optimization allows the algorithm to solve high-dimensional problems, and provides a low-dimensional representation which can be then be re-used. We evaluate the DDE approach by evolving solutions for inverse kinematics of a planar arm (200 joint angles) and for gaits of a 6-legged robot in action space (a sequence of 60 positions for each of the 12 joints). We show that the DDE approach not only accelerates and improves optimization, but produces a powerful encoding that captures a bias for high performance while expressing a variety of solutions.
The Covid-19 pandemic has challenged educators across the world to move their teaching and mentoring from in-person to remote. During nonpandemic semesters at their institutes (e.g. universities), educators can directly provide students the software environment needed to support their learning - either in specialized computer laboratories (e.g. computational chemistry labs) or shared computer spaces. These labs are often supported by staff that maintains the operating systems (OS) and software. But how does one provide a specialized software environment for remote teaching? One solution is to provide students a customized operating system (e.g., Linux) that includes open-source software for supporting your teaching goals. However, such a solution should not require students to install the OS alongside their existing one (i.e. dual/multi-booting) or be used as a complete replacement. Such approaches are risky because of a) the students' possible lack of software expertise, b) the possible disruption of an existing software workflow that is needed in other classes or by other family members, and c) the importance of maintaining a working computer when isolated (e.g. societal restrictions). To illustrate possible solutions, we discuss our approach that used a customized Linux OS and a Docker container in a course that teaches computational chemistry and Python3.
Background 3-hydroxy-3-methylglutaryl-coenzyme A lyase deficiency (HMGCLD) is an autosomal recessive disorder of ketogenesis and leucine degradation due to mutations in HMGCL .
Method We performed a systematic literature search to identify all published cases. 211 patients of whom relevant clinical data were available were included in this analysis. Clinical course, biochemical findings and mutation data are highlighted and discussed. An overview on all published HMGCL variants is provided.
Results More than 95% of patients presented with acute metabolic decompensation. Most patients manifested within the first year of life, 42.4% already neonatally. Very few individuals remained asymptomatic. The neurologic long-term outcome was favorable with 62.6% of patients showing normal development.
Conclusion This comprehensive data analysis provides a systematic overview on all published cases with HMGCLD including a list of all known HMGCL mutations.
2-methylacetoacetyl-coenzyme A thiolase (beta-ketothiolase) deficiency: one disease - two pathways
(2019)
Background: 2-methylacetoacetyl-coenzyme A thiolase deficiency (MATD; deficiency of mitochondrial acetoacetyl-coenzyme A thiolase T2/ “beta-ketothiolase”) is an autosomal recessive disorder of ketone body utilization and isoleucine degradation due to mutations in ACAT1.
Methods: We performed a systematic literature search for all available clinical descriptions of patients with MATD. 244 patients were identified and included in this analysis. Clinical course and biochemical data are presented and discussed.
Results: For 89.6 % of patients at least one acute metabolic decompensation was reported. Age at first symptoms ranged from 2 days to 8 years (median 12 months). More than 82% of patients presented in the first two years of life, while manifestation in the neonatal period was the exception (3.4%). 77.0% (157 of 204 patients) of patients showed normal psychomotor development without neurologic abnormalities.
Conclusion: This comprehensive data analysis provides a systematic overview on all cases with MATD identified in the literature. It demonstrates that MATD is a rather benign disorder with often favourable outcome, when compared with many other organic acidurias.
During the dawn of chemistry when the temperature of the young Universe had fallen below ∼4000 K, the ions of the light elements produced in Big Bang nucleosynthesis recombined in reverse order of their ionization potential. With its higher ionization potentials, He++ (54.5 eV) and He+ (24.6 eV) combined first with free electrons to form the first neutral atom, prior to the recombination of hydrogen (13.6 eV). At that time, in this metal-free and low-density environment, neutral helium atoms formed the Universe's first molecular bond in the helium hydride ion HeH+, by radiative association with protons (He + H+ → HeH+ + hν). As recombination progressed, the destruction of HeH+ (HeH+ + H → He + H+2) created a first path to the formation of molecular hydrogen, marking the beginning of the Molecular Age. Despite its unquestioned importance for the evolution of the early Universe, the HeH+ molecule has so far escaped unequivocal detection in interstellar space. In the laboratory, the ion was discovered as long ago as 1925, but only in the late seventies was the possibility that HeH+ might exist in local astrophysical plasmas discussed. In particular, the conditions in planetary nebulae were shown to be suitable for the production of potentially detectable HeH+ column densities: the hard radiation field from the central hot white dwarf creates overlapping Strömgren spheres, where HeH+ is predicted to form, primarily by radiative association of He+ and H. With the GREAT spectrometer onboard SOFIA, the HeH+ rotational ground-state transition at λ149.1 μm is now accessible. We report here its detection towards the planetary nebula NGC7027.
Describing the elephant: a foundational model of human needs, motivation, behaviour, and wellbeing
(2020)
Models of basic psychological needs have been present and popular in the academic and lay literature for more than a century yet reviews of needs models show an astonishing lack of consensus. This raises the question of what basic human psychological needs are and if this can be consolidated into a model or framework that can align previous research and empirical study. The authors argue that the lack of consensus arises from researchers describing parts of the proverbial elephant correctly but failing to describe the full elephant. Through redefining what human needs are and matching this to an evolutionary framework we can see broad consensus across needs models and neatly slot constructs and psychological and behavioural theories into this framework. This enables a descriptive model of drives, motives, and well-being that can be simply outlined but refined enough to do justice to the complexities of human behaviour. This also raises some issues of how subjective well-being is and should be measured. Further avenues of research and how to continue building this model and framework are proposed.
Quality diversity algorithms can be used to efficiently create a diverse set of solutions to inform engineers' intuition. But quality diversity is not efficient in very expensive problems, needing 100.000s of evaluations. Even with the assistance of surrogate models, quality diversity needs 100s or even 1000s of evaluations, which can make it use infeasible. In this study we try to tackle this problem by using a pre-optimization strategy on a lower-dimensional optimization problem and then map the solutions to a higher-dimensional case. For a use case to design buildings that minimize wind nuisance, we show that we can predict flow features around 3D buildings from 2D flow features around building footprints. For a diverse set of building designs, by sampling the space of 2D footprints with a quality diversity algorithm, a predictive model can be trained that is more accurate than when trained on a set of footprints that were selected with a space-filling algorithm like the Sobol sequence. Simulating only 16 buildings in 3D, a set of 1024 building designs with low predicted wind nuisance is created. We show that we can produce better machine learning models by producing training data with quality diversity instead of using common sampling techniques. The method can bootstrap generative design in a computationally expensive 3D domain and allow engineers to sweep the design space, understanding wind nuisance in early design phases.
In optimization methods that return diverse solution sets, three interpretations of diversity can be distinguished: multi-objective optimization which searches diversity in objective space, multimodal optimization which tries spreading out the solutions in genetic space, and quality diversity which performs diversity maintenance in phenotypic space. We introduce niching methods that provide more flexibility to the analysis of diversity and a simple domain to compare and provide insights about the paradigms. We show that multiobjective optimization does not always produce much diversity, quality diversity is not sensitive to genetic neutrality and creates the most diverse set of solutions, and multimodal optimization produces higher fitness solutions. An autoencoder is used to discover phenotypic features automatically, producing an even more diverse solution set. Finally, we make recommendations about when to use which approach.
In complex, expensive optimization domains we often narrowly focus on finding high performing solutions, instead of expanding our understanding of the domain itself. But what if we could quickly understand the complex behaviors that can emerge in said domains instead? We introduce surrogate-assisted phenotypic niching, a quality diversity algorithm which allows to discover a large, diverse set of behaviors by using computationally expensive phenotypic features. In this work we discover the types of air flow in a 2D fluid dynamics optimization problem. A fast GPU-based fluid dynamics solver is used in conjunction with surrogate models to accurately predict fluid characteristics from the shapes that produce the air flow. We show that these features can be modeled in a data-driven way while sampling to improve performance, rather than explicitly sampling to improve feature models. Our method can reduce the need to run an infeasibly large set of simulations while still being able to design a large diversity of air flows and the shapes that cause them. Discovering diversity of behaviors helps engineers to better understand expensive domains and their solutions.
The prototype of a workflow system for the submission of content to a digital object repository is here presented. It is based entirely on open-source standard components and features a service-oriented architecture. The front-end consists of Java Business Process Management (jBPM), Java Server Faces (JSF), and Java Server Pages (JSP). A Fedora Repository and a mySQL data base management system serve as a back-end. The communication between front-end and back-end uses a SOAP minimal binding stub. We describe the design principles and the construction of the prototype and discuss the possibilities and limitations of work ow creation by administrators. The code of the prototype is open-source and can be retrieved in the project escipub at http://sourceforge.net/ .
Since being introduced in the sixties and seventies, semi-implicit RosenbrockWanner (ROW) methods have become an important tool for the timeintegration of ODE and DAE problems. Over the years, these methods have been further developed in order to save computational effort by regarding approximations with respect to the given Jacobian [5], reduce effects of order reduction by introducing additional conditions [2, 4] or use advantages of partial explicit integration by considering underlying Runge-Kutta formulations [1]. As a consequence, there is a large number of different ROW-type schemes with characteristic properties for solving various problem formulations given in literature today.
Deep learning models are extensively used in various safety critical applications. Hence these models along with being accurate need to be highly reliable. One way of achieving this is by quantifying uncertainty. Bayesian methods for UQ have been extensively studied for Deep Learning models applied on images but have been less explored for 3D modalities such as point clouds often used for Robots and Autonomous Systems. In this work, we evaluate three uncertainty quantification methods namely Deep Ensembles, MC-Dropout and MC-DropConnect on the DarkNet21Seg 3D semantic segmentation model and comprehensively analyze the impact of various parameters such as number of models in ensembles or forward passes, and drop probability values, on task performance and uncertainty estimate quality. We find that Deep Ensembles outperforms other methods in both performance and uncertainty metrics. Deep ensembles outperform other methods by a margin of 2.4% in terms of mIOU, 1.3% in terms of accuracy, while providing reliable uncertainty for decision making.
Introduction: Antimicrobial resistance (AMR) has emerged as one of the leading threats to public health. AMR possesses a multidimensional challenge that has social, economic, and environmental dimensions that encompass the food production system, influencing human and animal health. The One Health approach highlights the inextricable linkage and interdependence between the health of people, animals, agriculture, and the environment. Antibiotic use in any of these One Health areas can potentially impact the health of other areas. There is a dearth of evidence on AMR from the natural environment, such as the plant-based agriculture sector. Antibiotics, antibiotic-resistant bacteria (ARB), and related AMR genes (ARGs) are assumed to present in the natural environment and disseminate resistance to fresh produce/vegetables and thus to human health upon consumption. Therefore, this study aims to investigate the role of vegetables in the spread of AMR through an agroecosystem exploration from a One Health perspective in Ahmedabad, India.
Protocol: The present study will be executed in Ahmedabad, located in Gujarat state in the Western part of India, by adopting a mixed-method approach. First, a systematic review will be conducted to document the prevalence of ARB and ARGs on fresh produce in South Asia. Second, agriculture farmland surveys will be used to collect the general farming practices and the data on common vegetables consumed raw by the households in Ahmedabad. Third, vegetable and soil samples will be collected from the selected agriculture farms and analyzed for the presence or absence of ARB and ARGs using standard microbiological and molecular methods.
Discussion: The analysis will help to understand the spread of ARB/ARGs through the agroecosystem. This is anticipated to provide an insight into the current state of ARB/ARGs contamination of fresh produce/vegetables and will assist in identifying the relevant strategies for effectively controlling and preventing the spread of AMR.
In an effort to assist researchers in choosing basis sets for quantum mechanical modeling of molecules (i.e. balancing calculation cost versus desired accuracy), we present a systematic study on the accuracy of computed conformational relative energies and their geometries in comparison to MP2/CBS and MP2/AV5Z data, respectively. In order to do so, we introduce a new nomenclature to unambiguously indicate how a CBS extrapolation was computed. Nineteen minima and transition states of buta-1,3-diene, propan-2-ol and the water dimer were optimized using forty-five different basis sets. Specifically, this includes one Pople (i.e. 6-31G(d)), eight Dunning (i.e. VXZ and AVXZ, X=2-5), twenty-five Jensen (i.e. pc-n, pcseg-n, aug-pcseg-n, pcSseg-n and aug-pcSseg-n, n=0-4) and nine Karlsruhe (e.g. def2-SV(P), def2-QZVPPD) basis sets. The molecules were chosen to represent both common and electronically diverse molecular systems. In comparison to MP2/CBS relative energies computed using the largest Jensen basis sets (i.e. n=2,3,4), the use of smaller sizes (n=0,1,2 and n=1,2,3) provides results that are within 0.11--0.24 and 0.09-0.16 kcal/mol. To practically guide researchers in their basis set choice, an equation is introduced that ranks basis sets based on a user-defined balance between their accuracy and calculation cost. Furthermore, we explain why the aug-pcseg-2, def2-TZVPPD and def2-TZVP basis sets are very suitable choices to balance speed and accuracy.
Differential-Algebraic Equations and Beyond: From Smooth to Nonsmooth Constrained Dynamical Systems
(2018)
The present article presents a summarizing view at differential-algebraic equations (DAEs) and analyzes how new application fields and corresponding mathematical models lead to innovations both in theory and in numerical analysis for this problem class. Recent numerical methods for nonsmooth dynamical systems subject to unilateral contact and friction illustrate the topicality of this development.
The promotion of sustainable packaging is part of the European Green Deal and plays a key role in the EU’s social and political strategy. One option is the use of renewable resources and biomass waste as raw materials for polymer production. Lignocellulose biomass from annual and perennial industrial crops and agricultural residues are a major source of polysaccharides, proteins, and lignin, and can also be used to obtain plant-based extracts and essential oils. Therefore, these biomasses are considered as potential substitute for fossil-based resources. Here, the status quo of bio-based polymers is discussed and evaluated in terms of properties related to packaging applications such as gas and water vapor permeability as well as mechanical properties. So far, their practical use is still restricted due to lower performance in fundamental packaging functions that directly influence food quality and safety, the length of shelf life and thus the amount of food waste. Besides bio-based polymers, this review focuses on plant extracts as active packaging agents. Incorporating extracts of herbs, flowers, trees, and their fruits is inevitable to achieve desired material properties that are capable to prolong the food shelf life. Finally, the adoption potential of packaging based on polymers from renewable resources is discussed from a bioeconomy perspective.
Application of underwater robots are on the rise, most of them are dependent on sonar for underwater vision, but the lack of strong perception capabilities limits them in this task. An important issue in sonar perception is matching image patches, which can enable other techniques like localization, change detection, and mapping. There is a rich literature for this problem in color images, but for acoustic images, it is lacking, due to the physics that produce these images. In this paper we improve on our previous results for this problem (Valdenegro-Toro et al, 2017), instead of modeling features manually, a Convolutional Neural Network (CNN) learns a similarity function and predicts if two input sonar images are similar or not. With the objective of improving the sonar image matching problem further, three state of the art CNN architectures are evaluated on the Marine Debris dataset, namely DenseNet, and VGG, with a siamese or two-channel architecture, and contrastive loss. To ensure a fair evaluation of each network, thorough hyper-parameter optimization is executed. We find that the best performing models are DenseNet Two-Channel network with 0.955 AUC, VGG-Siamese with contrastive loss at 0.949 AUC and DenseNet Siamese with 0.921 AUC. By ensembling the top performing DenseNet two-channel and DenseNet-Siamese models overall highest prediction accuracy obtained is 0.978 AUC, showing a large improvement over the 0.91 AUC in the state of the art.
Facial emotion recognition is the task to classify human emotions in face images. It is a difficult task due to high aleatoric uncertainty and visual ambiguity. A large part of the literature aims to show progress by increasing accuracy on this task, but this ignores the inherent uncertainty and ambiguity in the task. In this paper we show that Bayesian Neural Networks, as approximated using MC-Dropout, MC-DropConnect, or an Ensemble, are able to model the aleatoric uncertainty in facial emotion recognition, and produce output probabilities that are closer to what a human expects. We also show that calibration metrics show strange behaviors for this task, due to the multiple classes that can be considered correct, which motivates future work. We believe our work will motivate other researchers to move away from Classical and into Bayesian Neural Networks.
Transdermal therapeutic systems (TTS) represent an up-to-day medication applied to human skin, which consists of a drug-containing pressure-sensitive adhesive (PSA) and a flexible backing layer. The development of a reliable TTS requires precise knowledge of the viscoelastic tack behavior of PSA in terms of adhesion and detaching. Tailoring of a PSA can be achieved by altering the resin content or modifying the chemical properties of the macromolecules. In this study, three different resin content of two silicone-based PSA – non-amine compatible, and less tack, amine-compatible – were investigated with the help of recently developed RheoTack method to characterize the retraction speed dependent tack behavior for various geometries of the testing rods. The obtained force-retraction displacement-curves clearly depict the effect of the chemical structure as well as the resin content. Decreasing the resin content shifts the start of fibril fracture to larger deformations states and significantly enhances the stretchability of the fibrils. To compare various rod geometries precisely, the force-retraction displacement curves were normalized to account for effective contact areas. The flat and spherical rods led to completely different failure and tack behaviors. Furthermore, the adhesion formation between TTS with flexible backing layers and rods during the dwell phase happens in a different manner compared to rigid plates, in particular for flat rods, where maximum compression stresses occur at the edges and not uniformly over the cross-section. Thus, the approach to follow ASTM D2949 has to be reconsidered for tests of these materials.
Reinforcement learning (RL) algorithms should learn as much as possible about the environment but not the properties of the physics engines that generate the environment. There are multiple algorithms that solve the task in a physics engine based environment but there is no work done so far to understand if the RL algorithms can generalize across physics engines. In this work, we compare the generalization performance of various deep reinforcement learning algorithms on a variety of control tasks. Our results show that MuJoCo is the best engine to transfer the learning to other engines. On the other hand, none of the algorithms generalize when trained on PyBullet. We also found out that various algorithms have a promising generalizability if the effect of random seeds can be minimized on their performance.
Start-ups stehen im Wettbewerb um qualifizierte Mitarbeiter in starker Konkurrenz zu etablierten Unternehmen und Konzernen. Der Bedarf an Fachkräften (etwa Software-Entwicklern) ist größer als je zuvor [1]. Wie stellen sich Start-ups als Arbeitgeber dar, um Personal für sich zu gewinnen? Dieser Frage wurde im Rahmen der Studie „Start-ups als Arbeitgeber“ nachgegangen.
Die Bilder aus dem Silicon Valley sind bekannt: Das Großraumbüro mit den Sitzecken zum Zurückziehen. Schaukeln, Kickern und Videospiele zum Relaxen in den Arbeitspausen. Überall gibt es etwas zu essen und zu trinken, und das natürlich gratis. – Diese Vorstellungen haben viele im Kopf. Finden sich diese Bilder in der Selbstdarstellung deutscher Start-ups als Arbeitgeber wieder?
Die hier vorgestellte Studie will keine allgemeingültigen Ergebnisse liefern, sondern ist explorativ angelegt und soll zur weiteren Beschäftigung mit diesem Forschungsfeld in Wissenschaft und Praxis anregen.
Object detectors have improved considerably in the last years by using advanced CNN architectures. However, many detector hyper-parameters are generally manually tuned, or they are used with values set by the detector authors. Automatic Hyper-parameter optimization has not been explored in improving CNN-based object detectors hyper-parameters. In this work, we propose the use of Black-box optimization methods to tune the prior/default box scales in Faster R-CNN and SSD, using Bayesian Optimization, SMAC, and CMA-ES. We show that by tuning the input image size and prior box anchor scale on Faster R-CNN mAP increases by 2% on PASCAL VOC 2007, and by 3% with SSD. On the COCO dataset with SSD there are mAP improvement in the medium and large objects, but mAP decreases by 1% in small objects. We also perform a regression analysis to find the significant hyper-parameters to tune.
In this paper, we describe an approach that enables an autonomous system to infer the semantics of a command (i.e. a symbol sequence representing an action) in terms of the relations between changes in the observations and the action instances. We present a method of how to induce a theory (i.e. a semantic description) of the meaning of a command in terms of a minimal set of background knowledge. The only thing we have is a sequence of observations from which we extract what kinds of effects were caused by performing the command. This way, we yield a description of the semantics of the action and, hence, a definition.
Grasp verification is advantageous for autonomous manipulation robots as they provide the feedback required for higher level planning components about successful task completion. However, a major obstacle in doing grasp verification is sensor selection. In this paper, we propose a vision based grasp verification system using machine vision cameras, with the verification problem formulated as an image classification task. Machine vision cameras consist of a camera and a processing unit capable of on-board deep learning inference. The inference in these low-power hardware are done near the data source, reducing the robot's dependence on a centralized server, leading to reduced latency, and improved reliability. Machine vision cameras provide the deep learning inference capabilities using different neural accelerators. Although, it is not clear from the documentation of these cameras what is the effect of these neural accelerators on performance metrics such as latency and throughput. To systematically benchmark these machine vision cameras, we propose a parameterized model generator that generates end to end models of Convolutional Neural Networks(CNN). Using these generated models we benchmark latency and throughput of two machine vision cameras, JeVois A33 and Sipeed Maix Bit. Our experiments demonstrate that the selected machine vision camera and the deep learning models can robustly verify grasp with 97% per frame accuracy.
Long-term variability of solar irradiance and its implications for photovoltaic power in West Africa
(2020)
This paper addresses long-term changes in solar irradiance for West Africa (3° N to 20° N and 20° W to 16° E) and its implications for photovoltaic power systems. Here we use satellite irradiance (Surface Solar Radiation Data Set-Heliosat, Edition 2.1, SARAH-2.1) to derive photovoltaic yields. Based on 35 years of data (1983–2017) the temporal and regional variability as well as long-term trends of global and direct horizontal irradiance are analyzed. Furthermore, at four locations a detailed time series analysis is undertaken. The dry and the wet season are considered separately.
Saliency methods are frequently used to explain Deep Neural Network-based models. Adebayo et al.'s work on evaluating saliency methods for classification models illustrate certain explanation methods fail the model and data randomization tests. However, on extending the tests for various state of the art object detectors we illustrate that the ability to explain a model is more dependent on the model itself than the explanation method. We perform sanity checks for object detection and define new qualitative criteria to evaluate the saliency explanations, both for object classification and bounding box decisions, using Guided Backpropagation, Integrated Gradients, and their Smoothgrad versions, together with Faster R-CNN, SSD, and EfficientDet-D0, trained on COCO. In addition, the sensitivity of the explanation method to model parameters and data labels varies class-wise motivating to perform the sanity checks for each class. We find that EfficientDet-D0 is the most interpretable method independent of the saliency method, which passes the sanity checks with little problems.