arXiv subjects

Peter Salamon

Publications and source records attributed to Peter Salamon.

12 recordsLinked to original sources

arXiv2023

Undulations of smectic A layers in achiral liquid crystals manifested as stripe textures

Self-assembly of organic molecules represents a fascinating playground to create various liquid crystalline (LC) nanostructures. In this work, we study layer undulations on micrometer scale in smectic A phases for achiral compounds, experimentally demonstrated as regular stripe patterns induced by thermal treatment. Undulations including their anharmonic properties are evaluated by means of polarimetric imaging and light diffraction experiments in cells with various thicknesses. The key role in stripe formation is played by high negative values of the thermal expansion coefficient.

cond-mat.soft↗

arXiv2023

Estimating ensemble likelihoods for the Sentinel-1 based Global Flood Monitoring product of the Copernicus Emergency Management Service

The Global Flood Monitoring (GFM) system of the Copernicus Emergency Management Service (CEMS) addresses the challenges and impacts that are caused by flooding. The GFM system provides global, near-real time flood extent masks for each newly acquired Sentinel-1 Interferometric Wide Swath Synthetic Aperture Radar (SAR) image, as well as flood information from the whole Sentinel-1 archive from 2015 on. The GFM flood extent is an ensemble product based on a combination of three independently developed flood mapping algorithms that individually derive the flood information from Sentinel-1 data. Each flood algorithm also provides classification uncertainty information that is aggregated into the GFM ensemble likelihood product as the mean of the individual classification likelihoods. As the flood detection algorithms derive uncertainty information with different methods, the value range of the three input likelihoods must be harmonized to a range from low [0] to high [100] flood likelihood. The ensemble likelihood is evaluated on two test sites in Myanmar and Somalia, showcasing the performance during an actual flood event and an area with challenging conditions for SAR-based flood detection. The Myanmar use case demonstrates the robustness if flood detections in the ensemble step disagree and how that information is communicated to the end-user. The Somalia use case demonstrates a setting where misclassifications are likely, how the ensemble process mitigates false detections and how the flood likelihoods can be interpreted to use such results with adequate caution.

physics.geo-ph↗

arXiv2022

Estimating Classification Confidence Using Kernel Densities

This paper investigates the post-hoc calibration of confidence for "exploratory" machine learning classification problems. The difficulty in these problems stems from the continuing desire to push the boundaries of which categories have enough examples to generalize from when curating datasets, and confusion regarding the validity of those categories. We argue that for such problems the "one-versus-all" approach (top-label calibration) must be used rather than the "calibrate-the-full-response-matrix" approach advocated elsewhere in the literature. We introduce and test four new algorithms designed to handle the idiosyncrasies of category-specific confidence estimation. Chief among these methods is the use of kernel density ratios for confidence calibration including a novel, bulletproof algorithm for choosing the bandwidth. We test our claims and explore the limits of calibration on a bioinformatics application (PhANNs) as well as the classic MNIST benchmark. Finally, our analysis argues that post-hoc calibration should always be performed, should be based only on the test dataset, and should be sanity-checked visually.

stat.ML↗

arXiv2020

Social Media Alerts can Improve, but not Replace Hydrological Models for Forecasting Floods

Social media can be used for disaster risk reduction as a complement to traditional information sources, and the literature has suggested numerous ways to achieve this. In the case of floods, for instance, data collection from social media can be triggered by a severe weather forecast and/or a flood prediction. By way of contrast, in this paper we explore the possibility of having an entirely independent flood monitoring system which is based completely on social media, and which is completely self-activated. This independence and self-activation would bring increased robustness, as the system would not depend on other mechanisms for forecasting. We observe that social media can indeed help in the early detection of some flood events that would otherwise not be detected until later, albeit at the cost of many false positives. Overall, our experiments suggest that social media signals should only be used to complement existing monitoring systems, and we provide various explanations to support this argument.

cs.SI↗

arXiv2020

Quantum finite-time thermodynamics: insight from a single qubit engine

Incorporating time into thermodynamics allows addressing the tradeoff between efficiency and power. A qubit engine serves as a toy model to study this tradeoff from first principles, based on the quantum theory of open systems. We study the quantum origin of irreversibility, originating from heat transport, quantum friction and thermalization in the presence of external driving. We construct various finite-time engine cycles based on the Otto and Carnot templates. Our analysis highlights the role of coherence and the quantum origin of entropy production.

quant-ph↗

arXiv2019

Annihilation of Point Defect Pairs in Freely Suspended Liquid-Crystal Films

We study the annihilation of topological defect pairs in the quasi-twodimensional (2D) geometry of freely suspended smectic films. This elementary process is at the basis of all models describing the statistics of complex defect patterns. We prepare pairs with opposite topological charges and retrieve the interaction mechanisms from their trajectories. The square-root dependence of the defect separation on the time until annihilation and the asymmetry in propagation velocities of the opponents predicted by theory are confirmed. The importance of defect orientations is demonstrated. Trajectories are in general curved, depending on the mutual orientations (phase mismatch) of the defects and on the orientation of the pair respective to the far, undisturbed director. The experiments provide the basis for an adaption of the theoretical models to the real complexity of the annihilation.

cond-mat.soft↗

arXiv2018

Quantum heat engines: limit cycles and exceptional points

We show that the inability of a quantum Otto cycle to reach a limit cycle is connected with the propagator of the cycle being non-compact. For a working fluid consisting of quantum harmonic oscillators, the transition point in parameter space where this instability occurs is associated with a non-hermitian degeneracy (exceptional point) of the eigenvalues of the propagator. In particular, a third-order exceptional point is observed at the transition from the region where the eigenvalues are complex numbers to the region where all the eigenvalues are real. Within this region we find another exceptional point, this time of second order, at which the trajectory becomes divergent. The onset of the divergent behavior corresponds to the modulus of one of the eigenvalues becoming larger than one. The physical origin of this phenomenon is that the hot and cold heat baths are unable to dissipate the frictional internal heat generated in the adiabatic strokes of the cycle. This behavior is contrasted with that of quantum spins as working fluid which have a compact Hamiltonian and thus no exceptional points. All arguments are rigorously proved in terms of the systems' associated Lie algebras.

quant-ph↗

arXiv2014

Flashing flexodomains and electroconvection rolls in a nematic liquid crystal

Pattern forming instabilities induced by ultralow frequency sinusoidal voltages were studied in a rod-like nematic liquid crystal by microscopic observations and simultaneous electric current measurements. Two pattern morphologies, electroconvection (EC) and flexodomains (FD), were distinguished; both appearing as time separated flashes within each half period of driving. A correlation was found between the time instants of the EC flashes and that of the nonlinear current response. The voltage dependence of the pattern contrast C(U) for EC has a different character than that for the FD. The flattening of C(U) at reducing the frequency was described in terms of an imperfect bifurcation model. Analysing the threshold characteristics of FD the temperature dependence of the difference |e_1-e_3| of the flexoelectric coefficients were also determined by considering elastic anisotropy.

cond-mat.soft↗

arXiv2014

Inhibited pattern formation by asymmetrical high voltage excitation in nematic fluids

In contrast to the predictions of the standard theory of electroconvection (EC), our experiments showed that the action of superposed ac and dc voltages rather inhibits pattern formation than favours the emergence of instabilities; the patternless region may extend to much higher voltages than the individual ac or dc thresholds. The pattern formation induced by such asymmetrical voltage was explored in a nematic liquid crystal in a wide frequency range. The findings could be qualitatively explained for the conductive EC, but represent a challenging problem for the dielectric EC.

cond-mat.soft↗

arXiv2014

Patterns driven by combined AC and DC electric fields in nematic liquid crystals

The effect of superimposed ac and dc electric fields on the formation of electroconvection and flexoelectric patterns in nematic liquid crystals was studied. For selected ac frequencies an extended standard model of the electro-hydrodynamic instabilities was used to characterize the onset of pattern formation in the two-dimensional parameter space of the magnitudes of the ac and dc electric field components. Numerical as well as approximate analytical calculations demonstrate that depending on the type of patterns and on the ac frequency, the combined action of ac and dc fields may either enhance or suppress the formation of patterns. The theoretical predictions are qualitatively confirmed by experiments in most cases. Some discrepancies, however, seem to indicate the need to extend the theoretical description.

cond-mat.soft↗

arXiv2012

Controllability on relaxation-free subspaces: On the relationship between adiabatic population transfer and optimal control

We consider the optimal control problem of transferring population between states of a quantum system where the coupling proceeds only via intermediate states that are subject to decay. We pose the question whether it is generally possible to carry out this transfer. For a single intermediate decaying state, we recover the Stimulated Raman Adiabatic Passage (STIRAP) process which we identify as the global optimum in the limit of infinite control time. We also present analytic solutions for the case of transfer that has to proceed via two consecutive intermediate decaying states. We show that in this case, for finite power the optimal control does not approach perfect state transfer even in the infinite time limit. We generalize our findings to characterize the topologies of paths that can be achieved by coherent control under the assumption of finite power.

quant-ph↗

arXiv2008

The Quantum Refrigerator: The quest for absolute zero

The scaling of the optimal cooling power of a reciprocating quantum refrigerator is sought as a function of the cold bath temperature as $T_c \to 0$. The working medium consists of noninteracting particles in a harmonic potential. Two closed-form solutions of the refrigeration cycle are analyzed, and compared to a numerical optimization scheme, focusing on cooling toward zero temperature. The optimal cycle is characterized by linear relations between the heat extracted from the cold bath, the energy level spacing of the working medium and the temperature. The scaling of the optimal cooling rate is found to be proportional to $T_c^{3/2}$ giving a dynamical interpretation to the third law of thermodynamics.

quant-ph↗