Search

An engine producing a finite power at the ideal (Carnot) efficiency is a dream engine, which is not prohibited by the thermodynamic second law. Some years ago, a two-terminal heat engine with {\em asymmetric} Onsager coefficients in the linear response regime was suggested by Benenti, Saito, and Casati [Phys. Rev. Lett. {\bf 106}, 230602 (2011)], as a prototypical system to make such a dream come true with non-divergent system parameter values. However, such a system has never been realized in spite of many trials. Here, we introduce an exactly solvable two-terminal Brownian heat engine with the asymmetric Onsager coefficients in the presence of a Lorenz (magnetic) force. Nevertheless, we show that the dream engine regime cannot be accessible even with the asymmetric Onsager coefficients, due to an instability keeping the engine from reaching its steady state. This is consistent with recent trade-off relations between the engine power and efficiency, where the (cyclic) steady-state condition is implicitly presumed. We conclude that the inaccessibility to the dream engine originates from the steady-state constraint on the engine., Comment: 7 pages, 3 figures

We propose a neural network-based approach that computes a stable and generalizing metric (LSiM) to compare data from a variety of numerical simulation sources. We focus on scalar time-dependent 2D data that commonly arises from motion and transport-based partial differential equations (PDEs). Our method employs a Siamese network architecture that is motivated by the mathematical properties of a metric. We leverage a controllable data generation setup with PDE solvers to create increasingly different outputs from a reference simulation in a controlled environment. A central component of our learned metric is a specialized loss function that introduces knowledge about the correlation between single data samples into the training process. To demonstrate that the proposed approach outperforms existing metrics for vector spaces and other learned, image-based metrics, we evaluate the different methods on a large range of test data. Additionally, we analyze generalization benefits of an adjustable training data difficulty and demonstrate the robustness of LSiM via an evaluation on three real-world data sets., Comment: Main paper: 9 pages, Appendix: 19 pages. Accepted at ICML 2020. Source code available at https://github.com/tum-pbs/LSIM and further information at https://ge.in.tum.de/publications/2020-lsim-kohl/

We consider a collapsing shell of matter to form the Hayward black hole and investigate semiclassically quantum radiation from the shell. Using the Israel's formulation, we obtain the mass relation between the collapsing shell and the Hayward black hole. By using the functional Schr\"{o}dinger formulation for the massless quantum radiation, the evolution of a vacuum state for a scalar field is shown to be unitary. We find that the number of quanta at a low frequency decreases for a large length parameter characterizing the Hayward black hole. Moreover, in the limit of low frequency, the Hawking temperature can be read off from the occupation number of excited states when the shell approaches its own horizon., Comment: 18 pages, 3 figures, published version in Phys. Rev. D

In pattern recognition or machine learning, it is a very fundamental task to find nearest neighbors of a given point. All the methods for the task work basically by comparing the given point to all the points in the data set. That is why the computational cost increases with the number of data points. However, the human visual system seems to work in a different way. When the human visual system tries to find the neighbors of one point on a map, it directly focuses on the area around the point and actively searches the neighbors by looking or zooming in and out around the point. In this paper, we propose an innovative search method for nearest neighbors, which seems very similar to how human visual system works on the task.

The discrete KdV (dKdV) equation, the pinnacle of discrete integrability, is often thought to possess the singularity confinement property because it confines on an elementary quadrilateral. Here we investigate the singularity structure of the dKdV equation through reductions of the equation, obtained for initial conditions on a staircase with height 1, and show that it is much more subtle than one might assume. We first study the singularities for the mappings obtained after reduction and contrast these with the singularities that arise in non-integrable generalizations of those mappings. We then show that the so-called `express method' for obtaining dynamical degrees for second order mappings can be succesfully applied to all the higher order mappings we derived. Finally, we use the information obtained on the singularity structure of the reductions to describe an important subset of singularity patterns for the dKdV equation and we present an example of a non-confining pattern and explain why its existence does not contradict the integrability of the dKdV equation., Comment: 20 pages, 10 figures [revised version with 6 references added]

This paper proposes an effective emotion control method for an end-to-end text-to-speech (TTS) system. To flexibly control the distinct characteristic of a target emotion category, it is essential to determine embedding vectors representing the TTS input. We introduce an inter-to-intra emotional distance ratio algorithm to the embedding vectors that can minimize the distance to the target emotion category while maximizing its distance to the other emotion categories. To further enhance the expressiveness of a target speech, we also introduce an effective interpolation technique that enables the intensity of a target emotion to be gradually changed to that of neutral speech. Subjective evaluation results in terms of emotional expressiveness and controllability show the superiority of the proposed algorithm to the conventional methods., Comment: Submitted to ICASSP 2020

We developed a new and computationally simple local block-wise self attention based normal structures segmentation approach applied to head and neck computed tomography (CT) images. Our method uses the insight that normal organs exhibit regularity in their spatial location and inter-relation within images, which can be leveraged to simplify the computations required to aggregate feature information. We accomplish this by using local self attention blocks that pass information between each other to derive the attention map. We show that adding additional attention layers increases the contextual field and captures focused attention from relevant structures. We developed our approach using U-net and compared it against multiple state-of-the-art self attention methods. All models were trained on 48 internal headneck CT scans and tested on 48 CT scans from the external public domain database of computational anatomy dataset. Our method achieved the highest Dice similarity coefficient segmentation accuracy of 0.85$\pm$0.04, 0.86$\pm$0.04 for left and right parotid glands, 0.79$\pm$0.07 and 0.77$\pm$0.05 for left and right submandibular glands, 0.93$\pm$0.01 for mandible and 0.88$\pm$0.02 for the brain stem with the lowest increase of 66.7\% computing time per image and 0.15\% increase in model parameters compared with standard U-net. The best state-of-the-art method called point-wise spatial attention, achieved \textcolor{black}{comparable accuracy but with 516.7\% increase in computing time and 8.14\% increase in parameters compared with standard U-net.} Finally, we performed ablation tests and studied the impact of attention block size, overlap of the attention blocks, additional attention layers, and attention block placement on segmentation performance.

Comparative evaluation lies at the heart of science, and determining the accuracy of a computational method is crucial for evaluating its potential as well as for guiding future efforts. However, metrics that are typically used have inherent shortcomings when faced with the under-resolved solutions of real-world simulation problems. We show how to leverage crowd-sourced user studies in order to address the fundamental problems of widely used classical evaluation metrics. We demonstrate that such user studies, which inherently rely on the human visual system, yield a very robust metric and consistent answers for complex phenomena without any requirements for proficiency regarding the physics at hand. This holds even for cases away from convergence where traditional metrics often end up inconclusive results. More specifically, we evaluate results of different essentially non-oscillatory (ENO) schemes in different fluid flow settings. Our methodology represents a novel and practical approach for scientific evaluations that can give answers for previously unsolved problems., Comment: 29 pages, 18 figures, 2 tables, more information: https://ge.in.tum.de/publications/2019-spotdiff-um/

We study the gravitational wave (GW) production in extremely strong first order phase transitions where the latent heat density dominates the plasma energy density, $\alpha \gtrsim 1$. In such transitions, bubbles develop extremely thin and relativistic fluid configurations, resulting in strong shock waves after collisions. We first propose a strategy to understand the GW production in such a system by separating the problem into the propagation part and the collision part. Focusing on the former, we next develop an effective theory for the propagation of the relativistic fluid shells. Using this effective theory, we finally calculate the expected duration of the relativistic fluid configurations and discuss its implications to the GW production., Comment: 34 pages, 14 figures

Mobile Crowd-Sensing (MCS) has appeared as a prospective solution for large-scale data collection, leveraging built-in sensors and social applications in mobile devices that enables a variety of Internet of Things (IoT) services. However, the human involvement in MCS results in a high possibility for unintentionally contributing corrupted and falsified data or intentionally spreading disinformation for malevolent purposes, consequently undermining IoT services. Therefore, recruiting trustworthy contributors plays a crucial role in collecting high-quality data and providing better quality of services while minimizing the vulnerabilities and risks to MCS systems. In this article, a novel trust model called Experience-Reputation (E-R) is proposed for evaluating trust relationships between any two mobile device users in a MCS platform. To enable the E-R model, virtual interactions among the users are manipulated by considering an assessment of the quality of contributed data from such users. Based on these interactions, two indicators of trust called Experience and Reputation are calculated accordingly. By incorporating the Experience and Reputation trust indicators (TIs), trust relationships between the users are established, evaluated and maintained. Based on these trust relationships, a novel trust-based recruitment scheme is carried out for selecting the most trustworthy MCS users to contribute to data sensing tasks. In order to evaluate the performance and effectiveness of the proposed trust-based mechanism as well as the E-R trust model, we deploy several recruitment schemes in a MCS testbed which consists of both normal and malicious users. The results highlight the strength of the trust-based scheme as it delivers better quality for MCS services while being able to detect malicious users., Comment: Published in IEEE Transactions on Information Forensics and Security on March 2019. 15 pages, 9 figures, 68 references

The output randomness from a random number generator can be certified by observing the violation of quantum contextuality inequalities based on the Kochen-Specker theorem. Contextuality can be tested in a single quantum system, which significantly simplifies the experimental requirements to observe the violation comparing to the ones based on nonlocality tests. However, it is not yet resolved how to ensure compatibilities for sequential measurements that is required in contextuality tests. Here, we employ a modified Klyachko-Can-Binicio\u{g}lu-Shumovsky contextuality inequality, which can ease the strict compatibility requirement on measurements. On a trapped single \Ba ion system, we experimentally demonstrate violation of the contextuality inequality and realize self-testing quantum random number expansion by closing detection loopholes. We perform $1.29 \times 10^8$ trials of experiments and extract the randomness of $8.06 \times 10^5$ bits with a speed of 270 bits s$^{-1}$. Our demonstration paves the way for the practical high-speed spot-checking quantum random number expansion and other secure information processing applications., Comment: Main text: 12 pages, 5 figures, Supplementary Materials: 5 pages

We investigate a nonequilibrium phase transition in a dissipative and coherent quantum spin system using the quantum Langevin equation and mean-field theory. Recently, the quantum contact process (QCP) was theoretically investigated using the Rydberg antiblockade effect, in particular, when the Rydberg atoms were excited in s-states so that their interactions were regarded as being between the nearest neighbors. However, when the atoms are excited to d-states, the dipole-dipole interactions become effective, and long-range interactions must be considered. Here, we consider a quantum spin model with a long-range QCP, where the branching and coagulation processes are allowed not only for the nearest-neighbor pairs, but also for long-distance pairs, coherently and incoherently. Using the semiclassical approach, we show that the mean-field phase diagram of our long-range model is similar to that of the nearest-neighbor QCP, where the continuous (discontinuous) transition is found in the weak (strong) quantum regime. However, at the tricritical point, we find a new universality class, which was neither that of the QCP at the tricritical point nor that of the classical directed percolation model with long-range interactions. Implementation of the long-range QCP using interacting cold gases is discussed., Comment: 10 pages, 2 figures

Entropy production (EP) is known as a fundamental quantity for measuring the irreversibility of processes in thermal equilibrium and states far from equilibrium. In stochastic thermodynamics, the EP becomes more visible in terms of the probability density functions of the trajectories of a particle in the state space. Inspired by a previous result that complex networks can serve as state spaces, we consider a data packet transport problem on complex networks. Entropy is produced owing to the complexity of pathways as the packet travels back and forth between two nodes. The EPs are exactly enumerated along the shortest paths between every pair of nodes, and the functional form of the EP distribution is determined by extreme value analysis. The asymptote of the accumulated EP distribution is found to follow the Gumbel distribution.

Microscopic (pore-scale) properties of porous media affect and often determine their macroscopic (continuum- or Darcy-scale) counterparts. Understanding the relationship between processes on these two scales is essential to both the derivation of macroscopic models of, e.g., transport phenomena in natural porous media, and the design of novel materials, e.g., for energy storage. Most microscopic properties exhibit complex statistical correlations and geometric constraints, which presents challenges for the estimation of macroscopic quantities of interest (QoIs), e.g., in the context of global sensitivity analysis (GSA) of macroscopic QoIs with respect to microscopic material properties. We present a systematic way of building correlations into stochastic multiscale models through Bayesian networks. This allows us to construct the joint probability density function (PDF) of model parameters through causal relationships that emulate engineering processes, e.g., the design of hierarchical nanoporous materials. Such PDFs also serve as input for the forward propagation of parametric uncertainty; our findings indicate that the inclusion of causal relationships impacts predictions of macroscopic QoIs. To assess the impact of correlations and causal relationships between microscopic parameters on macroscopic material properties, we use a moment-independent GSA based on the differential mutual information. Our GSA accounts for the correlated inputs and complex non-Gaussian QoIs. The global sensitivity indices are used to rank the effect of uncertainty in microscopic parameters on macroscopic QoIs, to quantify the impact of causality on the multiscale model's predictions, and to provide physical interpretations of these results for hierarchical nanoporous materials., Comment: 23 pages, 11 figures and 5 tables