Showing papers by "Technische Universität Darmstadt published in 2013"

A Survey on Policy Search for Robotics

Abstract: Policy search is a subfield in reinforcement learning which focuses on finding good parameters for a given policy parametrization. It is well suited for robotics as it can cope with high-dimensional state and action spaces, one of the main challenges in robot learning. We review recent successes of both model-free and model-based policy search in robot learning.Model-free policy search is a general approach to learn policies based on sampled trajectories. We classify model-free methods based on their policy evaluation strategy, policy update strategy, and exploration strategy and present a unified view on existing algorithms. Learning a policy is often easier than learning an accurate forward model, and, hence, model-free methods are more frequently used in practice. However, for each sampled trajectory, it is necessary to interact with the robot, which can be time consuming and challenging in practice. Model-based policy search addresses this problem by first learning a simulator of the robot's dynamics from data. Subsequently, the simulator generates trajectories that are used for policy learning. For both model-free and model-based policy search methods, we review their respective properties and their applicability to robotic systems.

Electroluminescence in Single Layer MoS2

Abstract: We detect electroluminescence in single layer molybdenum disulfide (MoS2) field-effect transistors built on transparent glass substrates. By comparing the absorption, photoluminescence, and electroluminescence of the same MoS2 layer, we find that they all involve the same excited state at 1.8 eV. The electroluminescence has pronounced threshold behavior and is localized at the contacts. The results show that single layer MoS2, a direct band gap semiconductor, could be promising for novel optoelectronic devices, such as two-dimensional light detectors and emitters.

Resource theory of quantum states out of thermal equilibrium.

Abstract: The ideas of thermodynamics have proved fruitful in the setting of quantum information theory, in particular the notion that when the allowed transformations of a system are restricted, certain states of the system become useful resources with which one can prepare previously inaccessible states. The theory of entanglement is perhaps the best-known and most well-understood resource theory in this sense. Here, we return to the basic questions of thermodynamics using the formalism of resource theories developed in quantum information theory and show that the free energy of thermodynamics emerges naturally from the resource theory of energy-preserving transformations. Specifically, the free energy quantifies the amount of useful work which can be extracted from asymptotically many copies of a quantum system when using only reversible energy-preserving transformations and a thermal bath at fixed temperature. The free energy also quantifies the rate at which resource states can be reversibly interconverted asymptotically, provided that a sublinear amount of coherent superposition over energy levels is available, a situation analogous to the sublinear amount of classical communication required for entanglement dilution.

Identification of polymer types and additives in marine microplastic particles using pyrolysis-GC/MS and scanning electron microscopy

Abstract: Any assessment of plastic contamination in the marine environment requires knowledge of the polymer type and the additive content of microplastics. Sequential pyrolysis-gas chromatography coupled to mass spectrometry (Pyr-GC/MS) was applied to simultaneously identify polymer types of microplastic particles and associated organic plastic additives (OPAs). In addition, a scanning electron microscope equipped with an energy-dispersive X-ray microanalyser was used to identify the inorganic plastic additives (IPAs) contained in these particles. A total of ten particles, which were optically identified as potentially being plastics, were extracted from two sediment samples collected from Norderney, a North Sea island, by density separation in sodium chloride. The weights of these blue, white and transparent fragments varied between 10 and 350 μg. Polymer types were identified by comparing the resulting pyrograms with those obtained from the pyrolysis of selected standard polymers. The particles consisted of polyethylene (PE), polypropylene, polystyrene, polyamide, chlorinated PE and chlorosulfonated PE. The polymers contained diethylhexyl phthalate, dibutyl phthalate, diethyl phthalate, diisobutyl phthalate, dimethyl phthalate, benzaldehyde and 2,4-di-tert-butylphenol. Sequential Py-GC/MS was found to be an appropriate tool for identifying marine microplastics for polymer types and OPAs. The IPAs identified were titanium dioxide nanoparticles (TiO2-NPs), barium, sulphur and zinc. When polymer–TiO2 composites are degraded in the marine environment, TiO2-NPs are probably released. Thus, marine microplastics may act as a TiO2-NP source, which has not yet been considered.

Centrality dependence of pi, K, and p production in Pb-Pb collisions at root s(NN)=2.76 TeV

Abstract: In this paper measurements are presented of pi(+/-), K-+/-, p, and (p) over bar production at midrapidity (vertical bar y vertical bar < 0.5), in Pb-Pb collisions at root s(NN) = 2.76 TeV as a function of centrality. The measurement covers the transverse-momentum (p(T)) range from 100, 200, and 300 MeV/c up to 3, 3, and 4.6 GeV/c for pi, K, and p, respectively. The measured p(T) distributions and yields are compared to expectations based on hydrodynamic, thermal and recombination models. The spectral shapes of central collisions show a stronger radial flow than measured at lower energies, which can be described in hydrodynamic models. In peripheral collisions, the p(T) distributions are not well reproduced by hydrodynamic models. Ratios of integrated particle yields are found to be nearly independent of centrality. The yield of protons normalized to pions is a factor similar to 1.5 lower than the expectation from thermal models.

Temperature-Insensitive (K,Na)NbO3-Based Lead-Free Piezoactuator Ceramics

Abstract: The development of lead-free piezoceramics has attracted great interest because of growing environmental concerns. A polymorphic phase transition (PPT) has been utilized in the past to tailor piezoelectric properties in lead-free (K,Na)NbO3 (KNN)-based materials accepting the drawback of large temperature sensitivity. Here a material concept is reported, which yields an average piezoelectric coefficientd33 of about 300 pC/N and a high level of unipolar strain up to 0.16% at room temperature. Most intriguingly, field-induced strain varies less than 10% from room temperature to 175 °C. The temperature insensitivity of field-induced strain is rationalized using an electrostrictive coupling to polarization amplitude while the temperature-dependent piezoelectric coefficient is discussed using localized piezoresponse probed by piezoforce microscopy. This discovery opens a new development window for temperature-insensitive piezoelectric actuators despite the presence of a polymorphic phase transition around room temperature.

Room-temperature entanglement between single defect spins in diamond

Abstract: Entanglement is the central yet fleeting phenomenon of quantum physics. Once being considered a peculiar counter-intuitive property of quantum theory1, it has developed into the most central element of quantum technology. Consequently, there have been a number of experimental demonstrations of entanglement between photons2, atoms3, ions4 and solid-state systems such as spins or quantum dots5, 6, 7, superconducting circuits8, 9 and macroscopic diamond10. Here we experimentally demonstrate entanglement between two engineered single solid-state spin quantum bits (qubits) at ambient conditions. Photon emission of defect pairs reveals ground-state spin correlation. Entanglement (fidelity = 0.67±0.04) is proved by quantum state tomography. Moreover, the lifetime of electron spin entanglement is extended to milliseconds by entanglement swapping to nuclear spins. The experiments mark an important step towards a scalable room-temperature quantum device being of potential use in quantum information processing as well as metrology.

Probabilistic Movement Primitives

Abstract: Movement Primitives (MP) are a well-established approach for representing modular and re-usable robot movement generators. Many state-of-the-art robot learning successes are based MPs, due to their compact representation of the inherently continuous and high dimensional robot movements. A major goal in robot learning is to combine multiple MPs as building blocks in a modular control architecture to solve complex tasks. To this effect, a MP representation has to allow for blending between motions, adapting to altered task variables, and co-activating multiple MPs in parallel. We present a probabilistic formulation of the MP concept that maintains a distribution over trajectories. Our probabilistic approach allows for the derivation of new operations which are essential for implementing all aforementioned properties in one framework. In order to use such a trajectory distribution for robot movement control, we analytically derive a stochastic feedback controller which reproduces the given trajectory distribution. We evaluate and compare our approach to existing methods on several simulated as well as real robot scenarios.

Equation of state and neutron star properties constrained by nuclear physics and observation

Abstract: Microscopic calculations of neutron matter based on nuclear interactions derived from chiral effective field theory, combined with the recent observation of a 1.97 +/- 0.04 M-circle dot neutron star, constrain the equation of state of neutron-rich matter at sub-and supranuclear densities. We discuss in detail the allowed equations of state and the impact of our results on the structure of neutron stars, the crust-core transition density, and the nuclear symmetry energy. In particular, we show that the predicted range for neutron star radii is robust. For use in astrophysical simulations, we provide detailed numerical tables for a representative set of equations of state consistent with these constraints.

What is the contact angle of water on graphene

Abstract: Although experimental and theoretical studies have addressed the question of the wetting properties of graphene, the actual value of the contact angle of water on an isolated graphene monolayer remains unknown. While recent experimental literature indicates that the contact angle of water on graphite is in the range 90–95°, it has been suggested that the contact angle on graphene may either be as high as 127° or moderately enhanced in comparison with graphite. With the support of classical molecular dynamics simulations using empirical force-fields, we develop an argumentation to show that the value of 127° is an unrealistic estimate and that a value of the order of 95–100° should be expected. Our study establishes a connection between the variation of the work of adhesion of water on graphene-based surfaces and the interaction potential between individual water molecules and these surfaces. We show that a variation of the contact angle from 90° on graphite to 127° on graphene would imply that both of the...

Realization of spin gapless semiconductors: the Heusler compound Mn2CoAl.

Abstract: Recent studies have reported an interesting class of semiconductor materials that bridge the gap between semiconductors and half-metallic ferromagnets. These materials, called spin gapless semiconductors, exhibit a band gap in one of the spin channels and a zero band gap in the other and thus allow for tunable spin transport. Here, we report the first experimental verification of the spin gapless magnetic semiconductor Mn(2)CoAl, an inverse Heusler compound with a Curie temperature of 720 K and a magnetic moment of 2 μ(B). Below 300 K, the compound exhibits nearly temperature-independent conductivity, very low, temperature-independent carrier concentration, and a vanishing Seebeck coefficient. The anomalous Hall effect is comparatively low, which is explained by the symmetry properties of the Berry curvature. Mn(2) CoAl is not only suitable material for room temperature semiconductor spintronics, the robust spin polarization of the spin gapless semiconductors makes it very promising material for spintronics in general.

Centrality determination of Pb-Pb collisions at root s(NN)=2.76 TeV with ALICE

Abstract: This publication describes the methods used to measure the centrality of inelastic Pb-Pb collisions at a center-of-mass energy of 2.76 TeV per colliding nucleon pair with ALICE. The centrality is a key parameter in the study of the properties of QCD matter at extreme temperature and energy density, because it is directly related to the initial overlap region of the colliding nuclei. Geometrical properties of the collision, such as the number of participating nucleons and the number of binary nucleon-nucleon collisions, are deduced from a Glauber model with a sharp impact parameter selection and shown to be consistent with those extracted from the data. The centrality determination provides a tool to compare ALICE measurements with those of other experiments and with theoretical calculations.

Learning to select and generalize striking movements in robot table tennis

Abstract: Learning new motor tasks from physical interactions is an important goal for both robotics and machine learning. However, when moving beyond basic skills, most monolithic machine learning approaches fail to scale. For more complex skills, methods that are tailored for the domain of skill learning are needed. In this paper, we take the task of learning table tennis as an example and present a new framework that allows a robot to learn cooperative table tennis from physical interaction with a human. The robot first learns a set of elementary table tennis hitting movements from a human table tennis teacher by kinesthetic teach-in, which is compiled into a set of motor primitives represented by dynamical systems. The robot subsequently generalizes these movements to a wider range of situations using our mixture of motor primitives approach. The resulting policy enables the robot to select appropriate motor primitives as well as to generalize between them. Finally, the robot plays with a human table tennis partner and learns online to improve its behavior. We show that the resulting setup is capable of playing table tennis using an anthropomorphic robot arm.

Charged particle therapy--optimization, challenges and future directions.

Abstract: The use of charged particle therapy to control tumours non-invasively offers advantages over conventional radiotherapy. Protons and heavy ions deposit energy far more selectively than X-rays, allowing a higher local control of the tumour, a lower probability of damage to healthy tissue, low risk of complications and the chance for a rapid recovery after therapy. Charged particles are also useful for treating tumours located in areas that surround tissues that are radiosensitive and in anatomical sites where surgical access is limited. Current trial outcomes indicate that accelerated ions can potentially replace surgery for radical cancer treatments, which might be beneficial as the success of surgical cancer treatments are largely dependent on the expertise and experience of the surgeon and the location of the tumour. However, to date, only a small number of controlled randomized clinical trials have made comparisons between particle therapy and X-rays. Therefore, although the potential advantages are clear and supported by data, the cost:benefit ratio remains controversial. Research in medical physics and radiobiology is focusing on reducing the costs and increasing the benefits of this treatment.

Masses of exotic calcium isotopes pin down nuclear forces

Abstract: The properties of exotic nuclei on the verge of existence play a fundamental part in our understanding of nuclear interactions. Exceedingly neutron-rich nuclei become sensitive to new aspects of nuclear forces. Calcium, with its doubly magic isotopes (40)Ca and (48)Ca, is an ideal test for nuclear shell evolution, from the valley of stability to the limits of existence. With a closed proton shell, the calcium isotopes mark the frontier for calculations with three-nucleon forces from chiral effective field theory. Whereas predictions for the masses of (51)Ca and (52)Ca have been validated by direct measurements, it is an open question as to how nuclear masses evolve for heavier calcium isotopes. Here we report the mass determination of the exotic calcium isotopes (53)Ca and (54)Ca, using the multi-reflection time-of-flight mass spectrometer of ISOLTRAP at CERN. The measured masses unambiguously establish a prominent shell closure at neutron number N = 32, in excellent agreement with our theoretical calculations. These results increase our understanding of neutron-rich matter and pin down the subtle components of nuclear forces that are at the forefront of theoretical developments constrained by quantum chromodynamics.

Effective inter-component communication mapping in Android with Epicc: an essential step towards holistic security analysis

Abstract: Many threats present in smartphones are the result of interactions between application components, not just artifacts of single components. However, current techniques for identifying inter-application communication are ad hoc and do not scale to large numbers of applications. In this paper, we reduce the discovery of inter-component communication (ICC) in smartphones to an instance of the Interprocedural Distributive Environment (IDE) problem, and develop a sound static analysis technique targeted to the Android platform. We apply this analysis to 1,200 applications selected from the Play store and characterize the locations and substance of their ICC. Experiments show that full specifications for ICC can be identified for over 93% of ICC locations for the applications studied. Further the analysis scales well; analysis of each application took on average 113 seconds to complete. Epicc, the resulting tool, finds ICC vulnerabilities with far fewer false positives than the next best tool. In this way, we develop a scalable vehicle to extend current security analysis to entire collections of applications as well as the interfaces they export.

Stopped light and image storage by electromagnetically induced transparency up to the regime of one minute.

Abstract: The maximal storage duration is an important benchmark for memories. In quantized media, storage times are typically limited due to stochastic interactions with the environment. Also, optical memories based on electromagnetically induced transparency (EIT) suffer strongly from such decoherent effects. External magnetic control fields may reduce decoherence and increase EIT storage times considerably but also lead to complicated multilevel structures. These are hard to prepare perfectly in order to push storage times toward the theoretical limit, i.e., the population lifetime T(1). We present a self-learning evolutionary strategy to efficiently drive an EIT-based memory. By combination of the self-learning loop for optimized optical preparation and improved dynamical decoupling, we extend EIT storage times in a doped solid above 40 s. Moreover, we demonstrate storage of images by EIT for 1 min. These ultralong storage times set a new benchmark for EIT-based memories. The concepts are also applicable to other storage protocols.

Toward supervised anomaly detection

Abstract: Anomaly detection is being regarded as an unsupervised learning task as anomalies stem from adversarial or unlikely events with unknown distributions. However, the predictive performance of purely unsupervised anomaly detection often fails to match the required detection rates in many tasks and there exists a need for labeled data to guide the model generation. Our first contribution shows that classical semi-supervised approaches, originating from a supervised classifier, are inappropriate and hardly detect new and unknown anomalies. We argue that semi-supervised anomaly detection needs to ground on the unsupervised learning paradigm and devise a novel algorithm that meets this requirement. Although being intrinsically non-convex, we further show that the optimization problem has a convex equivalent under relatively mild assumptions. Additionally, we propose an active learning strategy to automatically filter candidates for labeling. In an empirical study on network intrusion detection data, we observe that the proposed learning methodology requires much less labeled data than the state-of-the-art, while achieving higher detection accuracies.

Systematic coarse-graining methods for soft matter simulations - a review

Abstract: Multiscale modelling of soft matter is an emerging field that has made rapid progress in the past decade. Several methods for systematic coarse-graining of molecular liquids and soft matter systems have been proposed in recent years. Herein, we review these methods and discuss a selected number of applications as well as limitations of the models and remaining challenges in developing representative and transferable pair potentials.

Equation of state and neutron star properties constrained by nuclear physics and observation

Abstract: Microscopic calculations of neutron matter based on nuclear interactions derived from chiral effective field theory, combined with the recent observation of a 1.97 +- 0.04 M_sun neutron star, constrain the equation of state of neutron-rich matter at sub- and supranuclear densities. We discuss in detail the allowed equations of state and the impact of our results on the structure of neutron stars, the crust-core transition density, and the nuclear symmetry energy. In particular, we show that the predicted range for neutron star radii is robust. For use in astrophysical simulations, we provide detailed numerical tables for a representative set of equations of state consistent with these constraints.

Envy on Facebook:a hidden threat to users' life satisfaction?

Abstract: The wealth of social information presented on Facebook is astound- ing. While these affordances allow users to keep up-to-date, they also produce a basis for social comparison and envy on an unprecedented scale. Even though envy may endanger users' life satisfaction and lead to platform avoidance, no study exists uncovering this dynamics. To close this gap, we build on responses of 584 Facebook users collected as part of two independent studies. In study 1, we explore the scale, scope, and nature of envy incidents triggered by Face- book. In study 2, the role of envy feelings is examined as a mediator between intensity of passive following on Facebook and users' life satisfaction. Con- firming full mediation, we demonstrate that passive following exacerbates envy feelings, which decrease life satisfaction. From a provider's perspective, our findings signal that users frequently perceive Facebook as a stressful environ- ment, which may, in the long-run, endanger platform sustainability.

Neutron matter at next-to-next-to-next-to-leading order in chiral effective field theory

Abstract: Neutron matter presents a unique system for chiral effective field theory because all many-body forces among neutrons are predicted to next-to-next-to-next-to-leading order (${\mathrm{N}}^{3}\mathrm{LO}$). We present the first complete ${\mathrm{N}}^{3}\mathrm{LO}$ calculation of the neutron matter energy. This includes the subleading three-nucleon forces for the first time and all leading four-nucleon forces. We find relatively large contributions from ${\mathrm{N}}^{3}\mathrm{LO}$ three-nucleon forces. Our results provide constraints for neutron-rich matter in astrophysics with controlled theoretical uncertainties.

K-S(0) and Lambda Production in Pb-Pb Collisions at root s(NN)=2: 76 TeV

Abstract: The ALICE measurement of K^0_S and {\Lambda} production at mid-rapidity in Pb-Pb collisions at sqrt(sNN) = 2.76 TeV is presented. The transverse momentum (pT) spectra are shown for several collision centrality intervals and in the pT range from 0.4 GeV/c (0.6 GeV/c for {\Lambda}) to 12 GeV/c. The pT dependence of the {\Lambda}/K^0_S ratios exhibits maxima in the vicinity of 3 GeV/c, and the positions of the maxima shift towards higher pT with increasing collision centrality. The magnitude of these maxima increases by almost a factor of three between most peripheral and most central Pb-Pb collisions. This baryon excess at intermediate pT is not observed in pp interactions at sqrt(s) = 0.9 TeV and at sqrt(s) = 7 TeV. Qualitatively, the baryon enhancement in heavy-ion collisions is expected from radial flow. However, the measured pT spectra above 2 GeV/c progressively decouple from hydrodynamical-model calculations. For higher values of pT, models that incorporate the influence of the medium on the fragmentation and hadronization processes describe qualitatively the pT dependence of the {\Lambda}/K^0_S ratio.

Bright laser-driven neutron source based on the relativistic transparency of solids

Abstract: Neutrons are unique particles to probe samples in many fields of research ranging from biology to material sciences to engineering and security applications. Access to bright, pulsed sources is currently limited to large accelerator facilities and there has been a growing need for compact sources over the recent years. Short pulse laser driven neutron sources could be a compact and relatively cheap way to produce neutrons with energies in excess of 10 MeV. For more than a decade experiments have tried to obtain neutron numbers sufficient for applications. Our recent experiments demonstrated an ion acceleration mechanism based on the concept of relativistic transparency. Using this new mechanism, we produced an intense beam of high energy (up to 170 MeV) deuterons directed into a Be converter to produce a forward peaked neutron flux with a record yield, on the order of ${10}^{10}\text{ }\text{ }\mathrm{n}/\mathrm{sr}$. We present results comparing the two acceleration mechanisms and the first short pulse laser generated neutron radiograph.

Long-range angular correlations of pi, K and p in p-Pb collisions at root s(NN)=5.02 TeV

Abstract: Angular correlations between unidentified charged trigger particles and various species of charged associated particles (unidentified particles, pions, kaons, protons and antiprotons) are measured by the ALICE detector in p-Pb collisions at a nucleon-nucleon centre-of-mass energy of 5.02 TeV in the transverse-momentum range 0.3 < p(T) < 4 GeV/c. The correlations expressed as associated yield per trigger particle are obtained in the pseudorapidity range vertical bar n(lab)vertical bar < 0.8. Fourier coefficients are extracted from the long-range correlations projected onto the azimuthal angle difference and studied as a function of p(T) and in intervals of event multiplicity. In high-multiplicity events, the second-order coefficient for protons, 4, is observed to be smaller than that for pions, v(2)(pi), up to about p(T) = 2 GeV/c. To reduce correlations due to jets, the per-trigger yield measured in low-multiplicity events is subtracted from that in high-multiplicity events. A two-ridge structure is obtained for all particle species. The Fourier decomposition of this structure shows that the second-order coefficients for pions and kaons are similar. The v(2)(p) is found to be smaller at low P-T and larger at higher p(T) than v(2)(pi), with a crossing occurring at about 2 GeV/c. This is qualitatively similar to the elliptic-flow pattern observed in heavy-ion collisions. A mass ordering effect at low transverse momenta is consistent with expectations from hydrodynamic model calculations assuming a collectively expanding system. (C) 2013 CERN. Published by Elsevier B.V. All rights reserved.

Performance of the ALICE VZERO system

Abstract: ALICE is an LHC experiment devoted to the study of strongly interacting matter in proton-proton, proton-nucleus and nucleus-nucleus collisions at ultra-relativistic energies. The ALICE VZERO system, made of two scintillator arrays at asymmetric positions, one on each side of the interaction point, plays a central role in ALICE. In addition to its core function as a trigger source, the VZERO system is used to monitor LHC beam conditions, to reject beam-induced backgrounds and to measure basic physics quantities such as luminosity, particle multiplicity, centrality and event plane direction in nucleus-nucleus collisions. After describing the VZERO system, this publication presents its performance over more than four years of operation at the LHC.

Polymer-derived SiCN and SiOC ceramics – structure and energetics at the nanoscale

Abstract: The complex nanodomain structure and energetic features of polymer-derived SiCN and SiOC ceramics have been investigated by multinuclear 1D and 2D solid-state NMR, micro-Raman, SAXS, XRD, HRTEM and oxidative high temperature oxide melt solution calorimetry, respectively. Structural models consistent with all these lines of evidence put emphasis on interconnected fractal domains stabilized by the mixed-bond-structure and interface bonding between nanodomains. The structural and thermodynamic role of hydrogen in the stabilization of interfacial bonding between the nanodomains in polysilylcarbodiimide-derived SiCN is reported as well.