Showing papers by "Johannes Kepler University of Linz published in 2015"

Fast and Accurate Deep Network Learning by Exponential Linear Units (ELUs)

Abstract: We introduce the "exponential linear unit" (ELU) which speeds up learning in deep neural networks and leads to higher classification accuracies. Like rectified linear units (ReLUs), leaky ReLUs (LReLUs) and parametrized ReLUs (PReLUs), ELUs alleviate the vanishing gradient problem via the identity for positive values. However, ELUs have improved learning characteristics compared to the units with other activation functions. In contrast to ReLUs, ELUs have negative values which allows them to push mean unit activations closer to zero like batch normalization but with lower computational complexity. Mean shifts toward zero speed up learning by bringing the normal gradient closer to the unit natural gradient because of a reduced bias shift effect. While LReLUs and PReLUs have negative values, too, they do not ensure a noise-robust deactivation state. ELUs saturate to a negative value with smaller inputs and thereby decrease the forward propagated variation and information. Therefore, ELUs code the degree of presence of particular phenomena in the input, while they do not quantitatively model the degree of their absence. In experiments, ELUs lead not only to faster learning, but also to significantly better generalization performance than ReLUs and LReLUs on networks with more than 5 layers. On CIFAR-100 ELUs networks significantly outperform ReLU networks with batch normalization while batch normalization does not improve ELU networks. ELU networks are among the top 10 reported CIFAR-10 results and yield the best published result on CIFAR-100, without resorting to multi-view evaluation or model averaging. On ImageNet, ELU networks considerably speed up learning compared to a ReLU network with the same architecture, obtaining less than 10% classification error for a single crop, single model network.

Prospects of Nanoscience with Nanocrystals

Abstract: Colloidal nanocrystals (NCs, i.e., crystalline nanoparticles) have become an important class of materials with great potential for applications ranging from medicine to electronic and optoelectronic devices. Today’s strong research focus on NCs has been prompted by the tremendous progress in their synthesis. Impressively narrow size distributions of just a few percent, rational shape-engineering, compositional modulation, electronic doping, and tailored surface chemistries are now feasible for a broad range of inorganic compounds. The performance of inorganic NC-based photovoltaic and light-emitting devices has become competitive to other state-of-the-art materials. Semiconductor NCs hold unique promise for near- and mid-infrared technologies, where very few semiconductor materials are available. On a purely fundamental side, new insights into NC growth, chemical transformations, and self-organization can be gained from rapidly progressing in situ characterization and direct imaging techniques. New phenom...

Flexible high power-per-weight perovskite solar cells with chromium oxide–metal contacts for improved stability in air

Abstract: Photovoltaic technology requires light-absorbing materials that are highly efficient, lightweight, low cost and stable during operation. Organolead halide perovskites constitute a highly promising class of materials, but suffer limited stability under ambient conditions without heavy and costly encapsulation. Here, we report ultrathin (3 μm), highly flexible perovskite solar cells with stabilized 12% efficiency and a power-per-weight as high as 23 W g(-1). To facilitate air-stable operation, we introduce a chromium oxide-chromium interlayer that effectively protects the metal top contacts from reactions with the perovskite. The use of a transparent polymer electrode treated with dimethylsulphoxide as the bottom layer allows the deposition-from solution at low temperature-of pinhole-free perovskite films at high yield on arbitrary substrates, including thin plastic foils. These ultra-lightweight solar cells are successfully used to power aviation models. Potential future applications include unmanned aerial vehicles-from airplanes to quadcopters and weather balloons-for environmental and industrial monitoring, rescue and emergency response, and tactical security applications.

msa: an R package for multiple sequence alignment.

Abstract: UNLABELLED Although the R platform and the add-on packages of the Bioconductor project are widely used in bioinformatics, the standard task of multiple sequence alignment has been neglected so far. The msa package, for the first time, provides a unified R interface to the popular multiple sequence alignment algorithms ClustalW, ClustalOmega and MUSCLE. The package requires no additional software and runs on all major platforms. Moreover, the msa package provides an R interface to the powerful package shade which allows for flexible and customizable plotting of multiple sequence alignments. AVAILABILITY AND IMPLEMENTATION msa is available via the Bioconductor project: http://bioconductor.org/packages/release/bioc/html/msa.html. Further information and the R code of the example presented in this paper are available at http://www.bioinf.jku.at/software/msa/.

Highly Efficient Luminescence of Cu(I) Compounds: Thermally Activated Delayed Fluorescence Combined with Short-Lived Phosphorescence

Abstract: Luminescent materials showing thermally activated delayed fluorescence (TADF) have gained high attractiveness as emitters in organic light emitting diodes (OLEDs) and other photonic applications. Nevertheless, even utilization of TADF can be further improved, introducing a novel concept. This is demonstrated by a new class of brightly luminescent low-cost Cu(I) compounds, for which the emission stems from both the lowest excited triplet T1 and singlet S1 state. At T = 300 K, these materials exhibit quantum yields of more than ΦPL = 90% at short emission decay times. About 80% of the emission intensity stems from the singlet due to TADF, but importantly, an additional 20% is contributed by the lower lying triplet state according to effective spin–orbit coupling (SOC). SOC induces also a relatively large zero-field splitting of the triplet being unusual for Cu(I) complexes. Thus, the overall emission decay time is distinctly reduced. Combined use of both decay paths opens novel photonic applications, in par...

A Wireless Charging System Applying Phase-Shift and Amplitude Control to Maximize Efficiency and Extractable Power

Abstract: Wireless power transfer (WPT) is an emerging technology with an increasing number of potential applications to transfer power from a transmitter to a mobile receiver over a relatively large air gap. However, its widespread application is hampered due to the relatively low efficiency of current Wireless power transfer (WPT) systems. This study presents a concept to maximize the efficiency as well as to increase the amount of extractable power of a WPT system operating in nonresonant operation. The proposed method is based on actively modifying the equivalent secondary-side load impedance by controlling the phase-shift of the active rectifier and its output voltage level. The presented hardware prototype represents a complete wireless charging system, including a dc–dc converter which is used to charge a battery at the output of the system. Experimental results are shown for the proposed concept in comparison to a conventional synchronous rectification approach. The presented optimization method clearly outperforms state-of-the-art solutions in terms of efficiency and extractable power.

Standards for dielectric elastomer transducers

Abstract: Dielectric elastomer transducers consist of thin electrically insulating elastomeric membranes coated on both sides with compliant electrodes. They are a promising electromechanically active polymer technology that may be used for actuators, strain sensors, and electrical generators that harvest mechanical energy. The rapid development of this field calls for the first standards, collecting guidelines on how to assess and compare the performance of materials and devices. This paper addresses this need, presenting standardized methods for material characterisation, device testing and performance measurement. These proposed standards are intended to have a general scope and a broad applicability to different material types and device configurations. Nevertheless, they also intentionally exclude some aspects where knowledge and/or consensus in the literature were deemed to be insufficient. This is a sign of a young and vital field, whose research development is expected to benefit from this effort towards standardisation.

Management Accounting in Small and Medium-Sized Enterprises: Current Knowledge and Avenues for Further Research

Abstract: Management accounting in small and medium-sized enterprises (SMEs) has been an issue of growing interest in the management accounting literature in recent years. However, published research is fragmented, spanning various fields such as accounting, small business and entrepreneurship, general management, and production and operations management. This paper aims to synthesize extant knowledge on management accounting in SMEs and provide concrete suggestions on how to proceed. We performed a systematic literature review, synthesized the results, and identified research gaps. Our findings show that usage of management accounting is not only lower but also different in SMEs compared to larger entities. Based on a comparison of SMEs, the review explains how environmental, staff, and organizational factors significantly influence the organization of management accounting in SMEs and that SME performance in general benefits from proper management accounting. We conclude with several concrete research q...

Risk management in SMEs: a systematic review of available evidence

Abstract: Purpose – The purpose of this paper is to provide a systematic literature review of available research evidence on risk management in small and medium-sized enterprises (SMEs). The authors aim to reveal ambiguities, gaps and contradictions in the literature, and to sketch avenues for further research. Design/methodology/approach – The authors follow the tenets of Tranfield et al. (2003) for conducting a systematic literature review. Following a key word search and an assessment of fit for this review, 27 papers were analyzed with respect to bibliographical information, research design and findings. Findings – The review identified various types of risks that may occur in SMEs. In addition, the publication analysis demonstrates the importance of a risk management process in SMEs and that the characteristics of SME owners have a significant impact on their business strategies. Research limitations/implications – Additional empirical research on risk identification, risk analysis, strategy implementation and...

Crossovers are associated with mutation and biased gene conversion at recombination hotspots

Abstract: Meiosis is a potentially important source of germline mutations, as sites of meiotic recombination experience recurrent double-strand breaks (DSBs) However, evidence for a local mutagenic effect of recombination from population sequence data has been equivocal, likely because mutation is only one of several forces shaping sequence variation By sequencing large numbers of single crossover molecules obtained from human sperm for two recombination hotspots, we find direct evidence that recombination is mutagenic: Crossovers carry more de novo mutations than nonrecombinant DNA molecules analyzed for the same donors and hotspots The observed mutations were primarily CG to TA transitions, with a higher frequency of transitions at CpG than non-CpGs sites This enrichment of mutations at CpG sites at hotspots could predominate in methylated regions involving frequent single-stranded DNA processing as part of DSB repair In addition, our data set provides evidence that GC alleles are preferentially transmitted during crossing over, opposing mutation, and shows that GC-biased gene conversion (gBGC) predominates over mutation in the sequence evolution of hotspots These findings are consistent with the idea that gBGC could be an adaptation to counteract the mutational load of recombination

Wearable magnetic field sensors for flexible electronics.

Abstract: Highly flexible bismuth Hall sensors on polymeric foils are fabricated, and the key optimization steps that are required to boost their sensitivity to the bulk value are identified. The sensor can be bent around the wrist or positioned on the finger to realize an interactive pointing device for wearable electronics. Furthermore, this technology is of great interest for the rapidly developing market of -eMobility, for optimization of eMotors and magnetic bearings.

Early performance of a miniaturized leadless cardiac pacemaker: the Micra Transcatheter Pacing Study

Abstract: Aims Permanent cardiac pacing is the only effective treatment for symptomatic bradycardia, but complications associated with conventional transvenous pacing systems are commonly related to the pacing lead and pocket. We describe the early performance of a novel self-contained miniaturized pacemaker. Methods and results Patients having Class I or II indication for VVI pacing underwent implantation of a Micra transcatheter pacing system, from the femoral vein and fixated in the right ventricle using four protractible nitinol tines. Prespecified objectives were >85% freedom from unanticipated serious adverse device events (safety) and <2 V 3-month mean pacing capture threshold at 0.24 ms pulse width (efficacy). Patients were implanted ( n = 140) from 23 centres in 11 countries (61% male, age 77.0 ± 10.2 years) for atrioventricular block (66%) or sinus node dysfunction (29%) indications. During mean follow-up of 1.9 ± 1.8 months, the safety endpoint was met with no unanticipated serious adverse device events. Thirty adverse events related to the system or procedure occurred, mostly due to transient dysrhythmias or femoral access complications. One pericardial effusion without tamponade occurred after 18 device deployments. In 60 patients followed to 3 months, mean pacing threshold was 0.51 ± 0.22 V, and no threshold was ≥2 V, meeting the efficacy endpoint ( P < 0.001). Average R-wave was 16.1 ± 5.2 mV and impedance was 650.7 ± 130 ohms. Conclusion Early assessment shows the transcatheter pacemaker can safely and effectively be applied. Long-term safety and benefit of the pacemaker will further be evaluated in the trial. Clinical Trial Registration ClinicalTrials.gov ID [NCT02004873][1]. [1]: /lookup/external-ref?link_type=CLINTRIALGOV&access_num=NCT02004873&atom=%2Fehj%2F36%2F37%2F2510.atom

Directional, passive liquid transport: the Texas horned lizard as a model for a biomimetic ‘liquid diode’

Abstract: Moisture-harvesting lizards such as the Texas horned lizard (Iguanidae: Phrynosoma cornutum) live in arid regions. Special skin adaptations enable them to access water sources such as moist sand and dew: their skin is capable of collecting and transporting water directionally by means of a capillary system between the scales. This fluid transport is passive, i.e. requires no external energy, and directs water preferentially towards the lizard's snout. We show that this phenomenon is based on geometric principles, namely on a periodic pattern of interconnected half-open capillary channels that narrow and widen. Following a biomimetic approach, we used these principles to develop a technical prototype design. Building upon the Young-Laplace equation, we derived a theoretical model for the local behaviour of the liquid in such capillaries. We present a global model for the penetration velocity validated by experimental data. Artificial surfaces designed in accordance with this model prevent liquid flow in one direction while sustaining it in the other. Such passive directional liquid transport could lead to process improvements and reduction of resources in many technical applications.

Hexagonal silicon realized

Abstract: Silicon, arguably the most important technological semiconductor, is predicted to exhibit a range of new and interesting properties when grown in the hexagonal crystal structure. To obtain pure hexagonal silicon is a great challenge because it naturally crystallizes in the cubic structure. Here, we demonstrate the fabrication of pure and stable hexagonal silicon evidenced by structural characterization. In our approach, we transfer the hexagonal crystal structure from a template hexagonal gallium phosphide nanowire to an epitaxially grown silicon shell, such that hexagonal silicon is formed. The typical ABABAB... stacking of the hexagonal structure is shown by aberration-corrected imaging in transmission electron microscopy. In addition, X-ray diffraction measurements show the high crystalline purity of the material. We show that this material is stable up to 9 GPa pressure. With this development, we open the way for exploring its optical, electrical, superconducting, and mechanical properties.

Spinodal nanodecomposition in semiconductors doped with transition metals

Abstract: This review presents the recent progress in computational materials design, experimental realization, and control methods of spinodal nanodecomposition under three- and two-dimensional crystal-growth conditions in spintronic materials, such as magnetically doped semiconductors. The computational description of nanodecomposition, performed by combining first-principles calculations with kinetic Monte Carlo simulations, is discussed together with extensive electron microscopy, synchrotron radiation, scanning probe, and ion beam methods that have been employed to visualize binodal and spinodal nanodecomposition (chemical phase separation) as well as nanoprecipitation (crystallographic phase separation) in a range of semiconductor compounds with a concentration of transition metal (TM) impurities beyond the solubility limit. The role of growth conditions, codoping by shallow impurities, kinetic barriers, and surface reactions in controlling the aggregation of magnetic cations is highlighted. According to theoretical simulations and experimental results the TM-rich regions appear in the form of either nanodots (the dairiseki phase) or nanocolumns (the konbu phase) buried in the host semiconductor. Particular attention is paid to Mn-doped group III arsenides and antimonides, TM-doped group III nitrides, Mn- and Fe-doped Ge, and Cr-doped group II chalcogenides, in which ferromagnetic features persisting up to above room temperature correlate with the presence of nanodecomposition and account for the application-relevant magneto-optical and magnetotransport properties of these compounds. Finally, it is pointed out that spinodal nanodecomposition can be viewed as a new class of bottom-up approach to nanofabrication.

Global warming potential of hydrogen and methane production from renewable electricity via power-to-gas technology

Abstract: Power-to-gas technology enables storage of surplus electricity from fluctuating renewable sources such as wind power or photovoltaics, by generating hydrogen (H2) via water electrolysis, with optional methane (CH4) synthesis from carbon dioxide (CO2) and H2; the advantage of the latter is that CH4 can be fed into existing gas infrastructure. This paper presents a life cycle assessment (LCA) of this technological concept, evaluating the main parameters influencing global warming potential (GWP) and primary energy demand. The conducted LCA of power-to-gas systems includes the production of H2 or CH4 from cradle to gate. Product utilization was not evaluated but considered qualitatively during interpretation. Material and energy balances were modeled using the LCA software GaBi 5 (PE International). The assessed impacts of H2 and CH4 from power-to-gas were compared to those of reference processes, such as steam reforming of natural gas and crude oil as well as natural gas extraction. Sensitivity analysis was used to evaluate the influence of the type of electricity source, the efficiency of the electrolyzer, and the type of CO2 source used for methanation. The ecological performance of both H2 and CH4 produced via power-to-gas strongly depends on the electricity generation source. The assessed impacts of H2 production are only improved if GWP of the utilized electricity does not exceed 190 g CO2 per kWh. Due to reduced efficiency, the assessed impacts of CH4 are higher than that of H2. Thus, the environmental break-even point for CH4 production is 113 g CO2 per kWh if utilized CO2 is treated as a waste product, and 73 g CO2 per kWh if the CO2 separation effort is included. Electricity mix of EU-27 countries is therefore not at all suitable as an input. Utilization of renewable H2 and CH4 in the industry or the transport sector offers substantial reduction potential in GWP and primary energy demand. H2 and CH4 production through power-to-gas with electricity from renewable sources, such as wind power or photovoltaics, offers substantial potential to reduce GWP and primary energy demand. However, the input of electricity predominately generated from fossil resources leads to a higher environmental impact of H2 and CH4 compared to fossil reference processes and is not recommended. As previously bound CO2 is re-emitted when CH4 is utilized for instance in vehicles, the type of CO2 source and the allocation method have a significant influence on overall ecological performance.

Local spectroscopy of moiré-induced electronic structure in gate-tunable twisted bilayer graphene

Abstract: Twisted bilayer graphene (tBLG) forms a quasicrystal whose structural and electronic properties depend on the angle of rotation between its layers. Here, we present a scanning tunneling microscopy study of gate-tunable tBLG devices supported by atomically smooth and chemically inert hexagonal boron nitride (BN). The high quality of these tBLG devices allows identification of coexisting moir\'e patterns and moir\'e super-superlattices produced by graphene-graphene and graphene-BN interlayer interactions. Furthermore, we examine additional tBLG spectroscopic features in the local density of states beyond the first van Hove singularity. Our experimental data are explained by a theory of moir\'e bands that incorporates ab initio calculations and confirms the strongly nonperturbative character of tBLG interlayer coupling in the small twist-angle regime.

Security challenges for medical devices

Abstract: Implantable devices, often dependent on software, save countless lives. But how secure are they?

Generalized smart evolving fuzzy systems

Abstract: In this paper, we propose a new methodology for learning evolving fuzzy systems (EFS) from data streams in terms of on-line regression/system identification problems. It comes with enhanced dynamic complexity reduction steps, acting on model components and on the input structure and by employing generalized fuzzy rules in arbitrarily rotated position. It is thus termed as Gen-Smart-EFS (GS-EFS), short for generalized smart evolving fuzzy systems. Equipped with a new projection concept for high-dimensional kernels onto one-dimensional fuzzy sets, our approach is able to provide equivalent conventional TS fuzzy systems with axis-parallel rules, thus maintaining interpretability when inferring new query samples. The on-line complexity reduction on rule level integrates a new merging concept based on a combined adjacency–homogeneity relation between two clusters (rules). On input structure level, complexity reduction is motivated by a combined statistical-geometric concept and acts in a smooth and soft manner by incrementally adapting feature weights: features may get smoothly out-weighted over time (\(\rightarrow\)soft on-line dimension reduction) but also may become reactivated at a later stage. Out-weighted features will contribute little to the rule evolution criterion, which prevents the generation of unnecessary rules and reduces over-fitting due to curse of dimensionality. The criterion relies on a newly developed re-scaled Mahalanobis distance measure for assuring monotonicity between feature weights and distance values. Gen-Smart-EFS will be evaluated based on high-dimensional real-world data (streaming) sets and compared with other well-known (evolving) fuzzy systems approaches. The results show improved accuracy with lower rule base complexity as well as smaller rule length when using Gen-Smart-EFS.

An imperceptible plastic electronic wrap.

Abstract: Extremely compliant sub-2-μm sensor films enable temperature mapping on complex 3D objects, like integrated circuits on printed circuit boards, food packages, and on human skin. In their stretchable form, these metal films withstand strains up to 275%. This imperceptible electronic foil technology platform offers new avenues for the design of complex, hybrid rigid-island stretchable-interconnect electronic devices such as RGB light-emitting diode (LED) strips that can be stretched and twisted without impairing their function.

Age, job identification, and entrepreneurial intention

Abstract: Purpose – The purpose of this paper is to examine how age and job identification affect entrepreneurial intention. Design/methodology/approach – The researchers draw on a representative sample of the Austrian adult workforce and apply binary logistic regression on entrepreneurial intention. Findings – The findings reveal that as employees age they are less inclined to act entrepreneurially, and that their entrepreneurial intention is lower the more they identify with their job. Whereas gender, education, and previous entrepreneurial experience matter, leadership and having entrepreneurial parents seem to have no impact on the entrepreneurial intention of employees. Research limitations/implications – Implications relate to a contingency perspective on entrepreneurial intention where the impact of age is exacerbated by stronger identification with the job. Practical implications – Practical implications include the need to account for different motivational backgrounds when addressing entrepreneurial employees of different ages. Societal implications include the need to adopt an age perspective to foster entrepreneurial intentions within established organizations. Originality/value – While the study corroborates and extends findings from entrepreneurial intention research, it contributes new empirical insights to the age and job-dependent contingency perspective on entrepreneurial intention

The mobility of single-file water molecules is governed by the number of H-bonds they may form with channel-lining residues.

Abstract: Channel geometry governs the unitary osmotic water channel permeability, pf, according to classical hydrodynamics. Yet, pf varies by several orders of magnitude for membrane channels with a constriction zone that is one water molecule in width and four to eight molecules in length. We show that both the pf of those channels and the diffusion coefficient of the single-file waters within them are determined by the number NH of residues in the channel wall that may form a hydrogen bond with the single-file waters. The logarithmic dependence of water diffusivity on NH is in line with the multiplicity of binding options at higher NH densities. We obtained high-precision pf values by (i) having measured the abundance of the reconstituted aquaporins in the vesicular membrane via fluorescence correlation spectroscopy and via high-speed atomic force microscopy, and (ii) having acquired the vesicular water efflux from scattered light intensities via our new adaptation of the Rayleigh-Gans-Debye equation.

What is a feature?: a qualitative study of features in industrial software product lines

Abstract: The notion of features is commonly used to describe the functional and non-functional characteristics of a system. In software product line engineering, features often become the prime entities of software reuse and are used to distinguish the individual products of a product line. Properly decomposing a product line into features, and correctly using features in all engineering phases, is core to the immediate and long-term success of such a system. Yet, although more than ten different definitions of the term feature exist, it is still a very abstract concept. Definitions lack concrete guidelines on how to use the notion of features in practice.To address this gap, we present a qualitative empirical study on actual feature usage in industry. Our study covers three large companies and an in-depth, contextualized analysis of 23 features, perceived by the interviewees as typical, atypical (outlier), good, or bad representatives of features. Using structured interviews, we investigate the rationales that lead to a feature's perception, and identify and analyze core characteristics (facets) of these features. Among others, we find that good features precisely describe customer-relevant functionality, while bad features primarily arise from rashly executed processes. Outlier features, serving unusual purposes, are necessary, but do not require the full engineering process of typical features.

High yield and ultrafast sources of electrically triggered entangled-photon pairs based on strain-tunable quantum dots.

Abstract: Triggered sources of entangled photon pairs are key components in most quantum communication protocols. For practical quantum applications, electrical triggering would allow the realization of compact and deterministic sources of entangled photons. Entangled-light-emitting-diodes based on semiconductor quantum dots are among the most promising sources that can potentially address this task. However, entangled-light-emitting-diodes are plagued by a source of randomness, which results in a very low probability of finding quantum dots with sufficiently small fine structure splitting for entangled-photon generation (∼10(-2)). Here we introduce strain-tunable entangled-light-emitting-diodes that exploit piezoelectric-induced strains to tune quantum dots for entangled-photon generation. We demonstrate that up to 30% of the quantum dots in strain-tunable entangled-light-emitting-diodes emit polarization-entangled photons. An entanglement fidelity as high as 0.83 is achieved with fast temporal post selection. Driven at high speed, that is 400 MHz, strain-tunable entangled-light-emitting-diodes emerge as promising devices for high data-rate quantum applications.

Boolector 2.0

Abstract: In this paper, we discuss the most important changes and new features introduced with version 2.0 of our SMT solver Boolector, which placed first in the QF_BV and QF_ABV tracks of the SMT competition 2014. We further outline some features and techniques that where not yet described in the context of Boolector.