A5073027221- Black Holes and Theoretical Physics
- Algebraic structures and combinatorial models
- Image and Signal Denoising Methods
- Advanced Image Processing Techniques
- Reinforcement Learning in Robotics
- Asthma and respiratory diseases
- Plasmonic and Surface Plasmon Research
- Gaussian Processes and Bayesian Inference
- Image Enhancement Techniques
- Model Reduction and Neural Networks
- Photonic and Optical Devices
- Quantum Chromodynamics and Particle Interactions
- Homotopy and Cohomology in Algebraic Topology
- Nonlinear Waves and Solitons
- Flow Measurement and Analysis
- Obsessive-Compulsive Spectrum Disorders
- Advanced Vision and Imaging
- Organic Electronics and Photovoltaics
- Advanced Data Compression Techniques
- Cellular Mechanics and Interactions
- Bone health and osteoporosis research
- Metamaterials and Metasurfaces Applications
- Photoacoustic and Ultrasonic Imaging
- Video Coding and Compression Technologies
- Noncommutative and Quantum Gravity Theories
University of Southern Denmark
2022
Huawei Technologies (Sweden)
2022
University of Würzburg
2022
Rutgers Sexual and Reproductive Health and Rights
2021
Harvard University
2016-2017
Rutgers, The State University of New Jersey
2016-2017
California Institute of Technology
2011-2013
University of California, Los Angeles
2011
Gachon University Gil Medical Center
2008
Seoul National University
2005
A hallmark of many, sometimes life-threatening, inflammatory diseases and disorders is vascular leakage. The extent severity leakage broadly mediated by the integrity endothelial cell (EC) monolayer, which in turn governed three major interactions: cell-cell cell-substrate contacts, soluble mediators, biomechanical forces. potentially critical but essentially uninvestigated component mediating these interactions stiffness substrate to monolayer adherent. Accordingly, we investigated...
Finding an optimal device structure in the vast combinatorial design space of freeform nanophotonic has been enormous challenge. In this study, we propose physics-informed reinforcement learning (PIRL) that combines adjoint-based method with to improve sample efficiency by order magnitude compared conventional and overcome issue local minima. To illustrate these advantages PIRL over other optimization algorithms, a family one-dimensional metasurface beam deflectors using PIRL, exceeding most...
This paper reviews the challenge on constrained high dynamic range (HDR) imaging that was part of New Trends in Image Restoration and Enhancement (NTIRE) workshop, held conjunction with CVPR 2022. manuscript focuses competition set-up, datasets, proposed methods their results. The aims at estimating an HDR image from multiple respective low (LDR) observations, which might suffer under-or over-exposed regions different sources noise. is composed two tracks emphasis fidelity complexity...
The increasing demand on a versatile high-performance metasurface requires freeform design method that can handle huge space, which is many orders of magnitude larger than conventional fixed-shape optical structures. In this work, we formulate the designing process one-dimensional Si beam deflectors as reinforcement learning problem to find their optimal structures consistently without requiring any prior data. During training, deep Q-network-based agent stochastically explores device space...
We define "BPS graphs" on punctured Riemann surfaces associated with A N −1 theories of class $$ \mathcal{S} . BPS graphs provide a bridge between two powerful frameworks for studying the spectrum states: spectral networks and quivers. They arise from degenerate at maximal intersections walls marginal stability Coulomb branch. While is ill-defined such intersections, graph captures useful basis elementary states. The topology encodes quiver, even higher-rank certain partial punctures. lead...
Ozone causes airway hyperresponsiveness (AHR) and pulmonary inflammation. Rho kinase (ROCK) is a key regulator of smooth muscle cell contraction inflammatory migration. To determine the contribution two ROCK isoforms ROCK1 ROCK2 to ozone-induced AHR, we exposed wild-type, ROCK1(+/-), ROCK2(+/-) mice air or ozone (2 ppm for 3 h) evaluated 24 h later. haploinsufficiency did not affect responsiveness in air-exposed but significantly reduced with greater reduction despite increased...
We introduce a new perspective and generalization of spectral networks for 4d $$ \mathcal{N} = 2 theories class S associated to Lie algebras \mathfrak{g} A n , D E6, E7. Spectral directly compute the BPS spectra 2d on surface defects coupled theories. algebraic interpretation these emerges naturally from our construction, leading description 2d-4d wall-crossing phenomena. Our construction also provides an efficient framework study In addition, we consider novel types with minuscule...
Abstract The optical properties of thin-film light emitting diodes (LEDs) are strongly dependent on their structures due to interference inside the devices. However, complexity design space grows exponentially with number parameters, making it challenging optimize multilayer LEDs rigorous electromagnetic simulations. In this work, we demonstrate an artificial neural network that can predict extraction efficiency organic LED structure in 30 ms, which is ∼10 3 times faster than simulation a...
Airway smooth muscle cells (ASMCs) are phenotypically regulated to exist in either a proliferative or contractile state. However, the influence of other airway structural cell types on ASMC phenotype is largely unknown. Although epithelial known drive ASM proliferation, their effects uncertain. In current study, we tested hypothesis that reduce ASMCs. To do so, measured force production by traction microscopy, gene and protein expression, as well calcium release Fura-2 ratiometric imaging....
We study vacua and BPS spectra of canonical surface defects class $$ \mathcal{S} theories in different decoupling limits using ADE spectral networks. In some regions the IR moduli spaces these 2d-4d systems, mixing between 2d 4d states is suppressed, spectrum becomes that a \mathcal{N} = (2, 2) theory. For limits, we identify describing with nonlinear sigma models coset have been previously studied. also certain cases where limit defect exhibits set appear to be entirely new. A detailed...
This paper proposes a fast algorithm that converts DCT coefficients into integer transform coefficients, for the domain transcoding from MPEG-x to H.264. For in same resolution, 8 /spl times/ are converted four 4/spl 4 by decomposing conversion matrix sparse ones. reduction of resolution half, we also propose an lower band coefficients. The matrices derived this require fewer computations than direct and conventional matrices, thus overall using proposed requires less computational complexity.
Abstract Aims: We compared the bone mineral density (BMD) of spinal and femur between patients with cervical cancer a control group. Methods: retrospectively analyzed BMD in 40 72 women. The women were treated surgery for benign disease. Results: Age, height, bodyweight body mass index not significantly different two groups. Total femoral was lower ( P = 0.006) than that In multiple regression analysis, statistically significant predictor positively correlated total femur. Conclusion:...
Electromagnetic (EM) simulation plays a crucial role in analyzing and designing devices with sub-wavelength scale structures such as solar cells, semiconductor devices, image sensors, future displays integrated photonic devices. Specifically, optics problems estimating device nanophotonic provide intriguing research topics far-reaching real world impact. Traditional algorithms for tasks require iteratively refining parameters through simulations, which often yield sub-optimal results due to...
Low-light and blurring issues are prevalent when capturing photos at night, often due to the use of long exposure address dim environments. Addressing these joint problems can be challenging error-prone if an end-to-end model is trained without incorporating appropriate physical model. In this paper, we introduce JUDE, a Deep Joint Unrolling for Deblurring Low-Light Image Enhancement, inspired by image Based on Retinex theory model, low-light blurry input iteratively deblurred decomposed,...
We consider the low-energy limit of two-dimensional theory on multiple M2-branes suspended between a flat M5-brane and curved M5-brane. argue that it is described by an $\mathcal{N}=(2,2)$ supersymmetric Landau-Ginzburg model with superpotential determined shape M5-branes, which flows in to Kazama-Suzuki coset model. provide evidence studying ground states BPS spectra systems.
This paper reviews the challenge on constrained high dynamic range (HDR) imaging that was part of New Trends in Image Restoration and Enhancement (NTIRE) workshop, held conjunction with CVPR 2022. manuscript focuses competition set-up, datasets, proposed methods their results. The aims at estimating an HDR image from multiple respective low (LDR) observations, which might suffer under- or over-exposed regions different sources noise. is composed two tracks emphasis fidelity complexity...
In this paper, we devise a distributional framework on actor-critic as solution to instability, action type restriction, and conflation between samples statistics. We propose new method that minimizes the Cram\'er distance with multi-step Bellman target distribution generated from novel Sample-Replacement algorithm denoted SR($\lambda$), which learns correct value under multiple operations. Parameterizing Gaussian Mixture Model further improves efficiency performance of method, name GMAC....