arXiv subjects

T. D. Nguyen

Publications and source records attributed to T. D. Nguyen.

7 recordsLinked to original sources

arXiv2020

Deep Neural Network (DNN) for Water/Fat Separation: Supervised Training, Unsupervised Training, and No Training

Purpose: To use a deep neural network (DNN) for solving the optimization problem of water/fat separation and to compare supervised and unsupervised training. Methods: The current T2*-IDEAL algorithm for solving fat/water separation is dependent on initialization. Recently, deep neural networks (DNN) have been proposed to solve fat/water separation without the need for suitable initialization. However, this approach requires supervised training of DNN (STD) using the reference fat/water separation images. Here we propose two novel DNN water/fat separation methods 1) unsupervised training of DNN (UTD) using the physical forward problem as the cost function during training, and 2) no-training of DNN (NTD) using physical cost and backpropagation to directly reconstruct a single dataset. The STD, UTD and NTD methods were compared with the reference T2*-IDEAL. Results: All DNN methods generated consistent water/fat separation results that agreed well with T2*-IDEAL under proper initialization. Conclusion: The water/fat separation problem can be solved using unsupervised deep neural networks.

eess.IV↗

arXiv2019

Specific Heat Signature of the Berezinskii-Kosterlitz-Thouless Transition in Ultrathin Superconducting Films

The Berezinskii-Kosterlitz-Thouless (BKT) transition is expected to have a clear signature on the specific heat. The singularity at the transition temperature $T_{BKT}$ is predicted to be immeasurable, and a broad non-universal peak is expected at $T>T_{BKT}$. Up to date this has not been observed in two-dimensional superconductors. We use a unique highly sensitive technique to measure the specific heat of ultrathin Pb films. We find that thick films exhibit a specific heat jump at $T_C$ that is consistent with BCS theory. As the film thickness is reduced below the superconducting coherence length and the systems enters the 2D limit the specific heat reveals BKT-like behavior. We discuss these observations in the framework of the continuous BCS-BKT crossover as a function of film thickness.

cond-mat.supr-con↗

arXiv2016

A Fast High-Voltage Switching Multiwire Proportional Chamber

A new experiment, called DeeMe, which is designed to search for $μ$-e conversions with a sensitivity of $\mathcal{O}(10^{-14})$, is in preparation at the Japan Proton Accelerator Research Complex (J-PARC). It utilizes a high-quality pulsed proton beam from the Rapid Cycling Synchrotron at J-PARC. The detector for DeeMe must tolerate large pulses of prompt charged particles whose instantaneous hit rate is as large as 70 GHz/mm$^2$ in a time width of 200 ns and detect a single electron that arrives with delayed timing on the order of microseconds. A special wire chamber has been developed with a new dynamic gain control technique that reduces space charge effects. In this paper, we detail the novel detector scheme and operation verification.

physics.ins-det↗

arXiv2015

Optimization techniques for multivariate least trimmed absolute deviation estimation

Given a dataset an outlier can be defined as an observation that it is unlikely to follow the statistical properties of the majority of the data. Computation of the location estimate of is fundamental in data analysis, and it is well known in statistics that classical methods, such as taking the sample average, can be greatly affected by the presence of outliers in the data. Using the median instead of the mean can partially resolve this issue but not completely. For the univariate case, a robust version of the median is the Least Trimmed Absolute Deviation (LTAD) robust estimator introduced in~\cite{Tableman1994}, which has desirable asymptotic properties such as robustness, consistently, high breakdown and normality. There are different generalizations of the LTAD for multivariate data, depending on the choice of norm. In~\cite{ChaPitZiou:2015} we present such a generalization using the Euclidean norm and propose a solution technique for the resulting combinatorial optimization problem, based on a necessary condition, that results in a highly convergent local search algorithm. In this subsequent work we use the $L^1$ norm to generalize the LTAD to higher dimensions, and show that the resulting mixed integer programming problem has an integral relaxation, after applying an appropriate data transformation. Moreover, we utilize the structure of the problem to show that the resulting LP's can be solved efficiently using a subgradient optimization approach. The robust statistical properties of the proposed estimator are verified by extensive computational results.

math.ST↗

arXiv2010

From antiferromagnetism to superconductivity in Fe 1+y(Te1-x,Sex) (0 < x < 0.20): a neutron powder diffraction analysis

The nuclear and magnetic structure of Fe1+y(Te1-x,Sex) (0 < x < 0.20) compounds was analyzed between 2 K and 300 K by means of Rietveld refinement of neutron powder diffraction data. Samples with x < 0.075 undergo a tetragonal to monoclinic phase transition at low temperature, whose critical temperature decreases with increasing Se content; this structural transition is strictly coupled to a long range antiferromagnetic ordering at the Fe site. Both the transition to a monoclinic phase and the long range antiferromagnetism are suppressed for 0.10 < x < 0.20. The onset of the structural and of the magnetic transition remains coincident with the increase of Se substitution. The low temperature monoclinic crystal structure has been revised. Superconductivity arises for x > 0.05, therefore a significant region where superconductivity and long range antiferromagnetism coexist is present in the pseudo-binary FeTe - FeSe phase diagram.

cond-mat.supr-con↗

arXiv2008

Thermal properties of SmFeAs(O1-xFx) as probe of the interplay between electrons and phonons

A comparative study of thermal properties of SmFeAsO, SmFeAs(O0.93F0.07) and SmFeAs(O0.85F0.15) samples is presented. Specific heat and thermal conductivity show clear evidences of the spin density wave (SDW) ordering below TSDW 135 K in undoped SmFeAsO. At low level of F-doping, SmFeAs(O0.93F0.07), SDW ordering is suppressed and superconducting features are not yet optimally developed in both specific heat and thermal conductivity. At optimal level of F-doping SmFeAs(O0.85F0.15) anomalies related to the superconducting transition are well noticeable. By a compared analysis of doped and undoped samples we conclude that, despite F-doping modifies definitely the electronic ground state, it does not substantially alter phonon and electron parameters, like phonon modes, Sommerfeld coefficient, electro-phonon coupling. The analysis of the thermal conductivity curves provides an evaluation of SDW and superconducting energy gap, showing that phonons can suitably probe features of electronic ground state.

cond-mat.supr-con↗

arXiv2007

Device-spectroscopy of magnetic field effects in a polyfluorene organic light-emitting diode

We perform charge-induced absorption and electroluminescence spectroscopy in a polyfluorene organic magnetoresistive device. Our experiments allow us to measure the singlet exciton, triplet exciton and polaron densities in a live device under an applied magnetic field, and to distinguish between three different models that were proposed to explain organic magnetoresistance. These models are based on different spin-dependent interactions, namely exciton formation, triplet exciton-polaron quenching and bipolaron formation. We show that the singlet exciton, triplet exciton and polaron densities and conductivity all increase with increasing magnetic field. Our data are inconsistent with the exciton formation and triplet-exciton polaron quenching models.

cond-mat.soft↗