Technische Universität Darmstadt

About: Technische Universität Darmstadt is a education organization based out in Darmstadt, Germany. It is known for research contribution in the topics: Computer science & Context (language use). The organization has 17316 authors who have published 40619 publications receiving 937916 citations. The organization is also known as: Darmstadt University of Technology & University of Darmstadt.

Context-Aware Proactive Content Caching With Service Differentiation in Wireless Networks

Abstract: Content caching in small base stations or wireless infostations is considered to be a suitable approach to improve the efficiency in wireless content delivery. Placing the optimal content into local caches is crucial due to storage limitations, but it requires knowledge about the content popularity distribution, which is often not available in advance. Moreover, local content popularity is subject to fluctuations, since mobile users with different interests connect to the caching entity over time. Which content a user prefers may depend on the user’s context. In this paper, we propose a novel algorithm for context-aware proactive caching. The algorithm learns context-specific content popularity online by regularly observing context information of connected users, updating the cache content and observing cache hits subsequently. We derive a sublinear regret bound, which characterizes the learning speed and proves that our algorithm converges to the optimal cache content placement strategy in terms of maximizing the number of cache hits. Furthermore, our algorithm supports service differentiation by allowing operators of caching entities to prioritize customer groups. Our numerical results confirm that our algorithm outperforms state-of-the-art algorithms in a real world data set, with an increase in the number of cache hits of at least 14%.

Nanoparticle Characterization: What to Measure?

Abstract: What to measure? is a key question in nanoscience, and it is not straightforward to address as different physicochemical properties define a nanoparticle sample. Most prominent among these properties are size, shape, surface charge, and porosity. Today researchers have an unprecedented variety of measurement techniques at their disposal to assign precise numerical values to those parameters. However, methods based on different physical principles probe different aspects, not only of the particles themselves, but also of their preparation history and their environment at the time of measurement. Understanding these connections can be of great value for interpreting characterization results and ultimately controlling the nanoparticle structure-function relationship. Here, the current techniques that enable the precise measurement of these fundamental nanoparticle properties are presented and their practical advantages and disadvantages are discussed. Some recommendations of how the physicochemical parameters of nanoparticles should be investigated and how to fully characterize these properties in different environments according to the intended nanoparticle use are proposed. The intention is to improve comparability of nanoparticle properties and performance to ensure the successful transfer of scientific knowledge to industrial real-world applications.

Studies on product distributions of iron and cobalt catalyzed Fischer–Tropsch synthesis

Abstract: The dependencies of hydrocarbon product distributions of iron and cobalt catalyzed Fischer–Tropsch synthesis on partial pressures of reactants have been studied. For cobalt catalysts particular attention has been focussed on the modification of distributions by secondary chain growth of readsorbed 1-alkenes while for iron catalysts secondary chain growth has been proved as negligible. The widely discussed concept of two superimposed Anderson–Schulz-Flory distributions has been applied for the representation of product distributions for both iron and cobalt catalysts. Based on 1-alkene and ethene cofeeding experiments with cobalt catalysts and the promoter effect of alkali on iron catalysts the conclusion has been drawn that superimposed distributions with different chain growth probabilities are the result of different chain growth mechanisms. These results and the dependence of product distribution on the partial pressures of hydrogen and carbon monoxide have lead to the conjecture that products associated with the lower growth probability are formed by the well accepted CH2 insertion mechanism.

Chiral Three-Nucleon Interactions in Light Nuclei, Neutron-α Scattering, and Neutron Matter

Abstract: We present quantum Monte Carlo calculations of light nuclei, neutron-$\ensuremath{\alpha}$ scattering, and neutron matter using local two- and three-nucleon ($3N$) interactions derived from chiral effective field theory up to next-to-next-to-leading order (${\mathrm{N}}^{2}\mathrm{LO}$). The two undetermined $3N$ low-energy couplings are fit to the $^{4}\mathrm{He}$ binding energy and, for the first time, to the spin-orbit splitting in the neutron-$\ensuremath{\alpha}$ $P$-wave phase shifts. Furthermore, we investigate different choices of local $3N$-operator structures and find that chiral interactions at ${\mathrm{N}}^{2}\mathrm{LO}$ are able to simultaneously reproduce the properties of $A=3,4,5$ systems and of neutron matter, in contrast to commonly used phenomenological $3N$ interactions.

Chameleon: A Hybrid Secure Computation Framework for Machine Learning Applications

Abstract: We present Chameleon, a novel hybrid (mixed-protocol) framework for secure function evaluation (SFE) which enables two parties to jointly compute a function without disclosing their private inputs. Chameleon combines the best aspects of generic SFE protocols with the ones that are based upon additive secret sharing. In particular, the framework performs linear operations in the ring $\mathbb{Z}_{2^l}$ using additively secret shared values and nonlinear operations using Yao's Garbled Circuits or the Goldreich-Micali-Wigderson protocol. Chameleon departs from the common assumption of additive or linear secret sharing models where three or more parties need to communicate in the online phase: the framework allows two parties with private inputs to communicate in the online phase under the assumption of a third node generating correlated randomness in an offline phase. Almost all of the heavy cryptographic operations are precomputed in an offline phase which substantially reduces the communication overhead. Chameleon is both scalable and significantly more efficient than the ABY framework (NDSS'15) it is based on. Our framework supports signed fixed-point numbers. In particular, Chameleon's vector dot product of signed fixed-point numbers improves the efficiency of mining and classification of encrypted data for algorithms based upon heavy matrix multiplications. Our evaluation of Chameleon on a 5 layer convolutional deep neural network shows 133x and 4.2x faster executions than Microsoft CryptoNets (ICML'16) and MiniONN (CCS'17), respectively.