arXiv subjects

Jiahao Zhang

Publications and source records attributed to Jiahao Zhang.

At least 163 records · Page 9Linked to original sources

arXiv2022

Analyzing Green View Index and Green View Index best path using Google Street View and deep learning

As an important part of urban landscape research, analyzing and studying street-level greenery can increase the understanding of a city's greenery, contributing to better urban living environment planning and design. Planning the best path of urban greenery is a means to effectively maximize the use of urban greenery, which plays a positive role in the physical and mental health of urban residents and the path planning of visitors. In this paper, we used Google Street View (GSV) to obtain street view images of Osaka City. The semantic segmentation model is adopted to segment the street view images and analyze the Green View Index (GVI) of Osaka City. Based on the GVI, we take advantage of the adjacency matrix and Floyd-Warshall Algorithm to calculate Green View Index best path, solving the limitations of ArcGIS software. Our analysis not only allows the calculation of specific routes for the GVI best paths but also realizes the visualization and integration of neighborhood urban greenery. By summarizing all the data, we can conduct an intuitive feeling and objective analysis of the street-level greenery in the research area. Based on this, such as urban residents and visitors can maximize the available natural resources for a better life. The dataset and code are available at https://github.com/Jackieam/GVI-Best-Path.

cs.CV↗

arXiv2022

ALL-MASK: A Reconfigurable Logic Locking Method for Multicore Architecture with Sequential-Instruction-Oriented Key

Intellectual property (IP) piracy has become a non-negligible problem as the integrated circuit (IC) production supply chain is becoming increasingly globalized and separated that enables attacks by potentially untrusted attackers. Logic locking is a widely adopted method to lock the circuit module with a key and prevent hackers from cracking it. The key is the critical aspect of logic locking, but the existing works have overlooked three possible challenges of the key: safety of key storage, easy key-attempt from interface and key-related overheads, bringing the further challenges of low error rate and small state space. In this work, the key is dynamically generated by utilizing the huge space of a CPU core, and the unlocking is performed implicitly through the interconnection inside the chip. A novel low-cost logic reconfigurable gate is together proposed with ferroelectric FET (FeFET) to mitigate the reverse engineering and removal attack. Compared to the common logic locking methods, our proposed approach is 19,945 times more time consuming to traverse all the possible combinations in only 9-bit-key condition. Furthermore, our technique let key length increases this complexity exponentially and ensure the logic obfuscation effect.

cs.AR↗

arXiv2022

RMFGP: Rotated Multi-fidelity Gaussian process with Dimension Reduction for High-dimensional Uncertainty Quantification

Multi-fidelity modelling arises in many situations in computational science and engineering world. It enables accurate inference even when only a small set of accurate data is available. Those data often come from a high-fidelity model, which is computationally expensive. By combining the realizations of the high-fidelity model with one or more low-fidelity models, the multi-fidelity method can make accurate predictions of quantities of interest. This paper proposes a new dimension reduction framework based on rotated multi-fidelity Gaussian process regression and a Bayesian active learning scheme when the available precise observations are insufficient. By drawing samples from the trained rotated multi-fidelity model, the so-called supervised dimension reduction problems can be solved following the idea of the sliced average variance estimation (SAVE) method combined with a Gaussian process regression dimension reduction technique. This general framework we develop can effectively solve high-dimensional problems while the data are insufficient for applying traditional dimension reduction methods. Moreover, a more accurate surrogate Gaussian process model of the original problem can be obtained based on our trained model. The effectiveness of the proposed rotated multi-fidelity Gaussian process(RMFGP) model is demonstrated in four numerical examples. The results show that our method has better performance in all cases and uncertainty propagation analysis is performed for last two cases involving stochastic partial differential equations.

stat.ML↗

arXiv2022

MultiAuto-DeepONet: A Multi-resolution Autoencoder DeepONet for Nonlinear Dimension Reduction, Uncertainty Quantification and Operator Learning of Forward and Inverse Stochastic Problems

A new data-driven method for operator learning of stochastic differential equations(SDE) is proposed in this paper. The central goal is to solve forward and inverse stochastic problems more effectively using limited data. Deep operator network(DeepONet) has been proposed recently for operator learning. Compared to other neural networks to learn functions, it aims at the problem of learning nonlinear operators. However, it can be challenging by using the original model to learn nonlinear operators for high-dimensional stochastic problems. We propose a new multi-resolution autoencoder DeepONet model referred to as MultiAuto-DeepONet to deal with this difficulty with the aid of convolutional autoencoder. The encoder part of the network is designed to reduce the dimensionality as well as discover the hidden features of high-dimensional stochastic inputs. The decoder is designed to have a special structure, i.e. in the form of DeepONet. The first DeepONet in decoder is designed to reconstruct the input function involving randomness while the second one is used to approximate the solution of desired equations. Those two DeepONets has a common branch net and two independent trunk nets. This architecture enables us to deal with multi-resolution inputs naturally. By adding $L_1$ regularization to our network, we found the outputs from the branch net and two trunk nets all have sparse structures. This reduces the number of trainable parameters in the neural network thus making the model more efficient. Finally, we conduct several numerical experiments to illustrate the effectiveness of our proposed MultiAuto-DeepONet model with uncertainty quantification.

stat.ML↗

arXiv2022

PAGP: A physics-assisted Gaussian process framework with active learning for forward and inverse problems of partial differential equations

In this work, a Gaussian process regression(GPR) model incorporated with given physical information in partial differential equations(PDEs) is developed: physics-assisted Gaussian processes(PAGP). The targets of this model can be divided into two types of problem: finding solutions or discovering unknown coefficients of given PDEs with initial and boundary conditions. We introduce three different models: continuous time, discrete time and hybrid models. The given physical information is integrated into Gaussian process model through our designed GP loss functions. Three types of loss function are provided in this paper based on two different approaches to train the standard GP model. The first part of the paper introduces the continuous time model which treats temporal domain the same as spatial domain. The unknown coefficients in given PDEs can be jointly learned with GP hyper-parameters by minimizing the designed loss function. In the discrete time models, we first choose a time discretization scheme to discretize the temporal domain. Then the PAGP model is applied at each time step together with the scheme to approximate PDE solutions at given test points of final time. To discover unknown coefficients in this setting, observations at two specific time are needed and a mixed mean square error function is constructed to obtain the optimal coefficients. In the last part, a novel hybrid model combining the continuous and discrete time models is presented. It merges the flexibility of continuous time model and the accuracy of the discrete time model. The performance of choosing different models with different GP loss functions is also discussed. The effectiveness of the proposed PAGP methods is illustrated in our numerical section.

stat.ML↗

arXiv2022

Spatiotemporal differentiators generating optical vortices with transverse orbital angular momentum and detecting sharp change of pulse envelope

As a new degree of freedom for optical manipulation, recently spatiotemporal optical vortices (STOVs) carrying transverse orbital angular momentums have been experimentally demonstrated with pulse shapers. Here a spatiotemporal differentiator is proposed to generate STOVs with transverse orbital angular momentum. In order to create phase singularity in the spatiotemporal domain, the spatiotemporal differentiator is designed by breaking spatial mirror symmetry. In contrast to pulse shapers, the device proposed here is a simple one-dimensional periodic nanostructure and thus it is much more compact. For a normal incident pulse, the differentiator generates a transmitted STOV pulse with transverse orbital angular momentum. Furthermore, the interference of the generated STOVs can be used to detect the sharp changes of pulse envelopes, in both spatial and temporal dimensions.

physics.optics↗

arXiv2021

LMR-CBT: Learning Modality-fused Representations with CB-Transformer for Multimodal Emotion Recognition from Unaligned Multimodal Sequences

Learning modality-fused representations and processing unaligned multimodal sequences are meaningful and challenging in multimodal emotion recognition. Existing approaches use directional pairwise attention or a message hub to fuse language, visual, and audio modalities. However, those approaches introduce information redundancy when fusing features and are inefficient without considering the complementarity of modalities. In this paper, we propose an efficient neural network to learn modality-fused representations with CB-Transformer (LMR-CBT) for multimodal emotion recognition from unaligned multimodal sequences. Specifically, we first perform feature extraction for the three modalities respectively to obtain the local structure of the sequences. Then, we design a novel transformer with cross-modal blocks (CB-Transformer) that enables complementary learning of different modalities, mainly divided into local temporal learning,cross-modal feature fusion and global self-attention representations. In addition, we splice the fused features with the original features to classify the emotions of the sequences. Finally, we conduct word-aligned and unaligned experiments on three challenging datasets, IEMOCAP, CMU-MOSI, and CMU-MOSEI. The experimental results show the superiority and efficiency of our proposed method in both settings. Compared with the mainstream methods, our approach reaches the state-of-the-art with a minimum number of parameters.

cs.CV↗

arXiv2021

EvoGAN: An Evolutionary Computation Assisted GAN

The image synthesis technique is relatively well established which can generate facial images that are indistinguishable even by human beings. However, all of these approaches uses gradients to condition the output, resulting in the outputting the same image with the same input. Also, they can only generate images with basic expression or mimic an expression instead of generating compound expression. In real life, however, human expressions are of great diversity and complexity. In this paper, we propose an evolutionary algorithm (EA) assisted GAN, named EvoGAN, to generate various compound expressions with any accurate target compound expression. EvoGAN uses an EA to search target results in the data distribution learned by GAN. Specifically, we use the Facial Action Coding System (FACS) as the encoding of an EA and use a pre-trained GAN to generate human facial images, and then use a pre-trained classifier to recognize the expression composition of the synthesized images as the fitness function to guide the search of the EA. Combined random searching algorithm, various images with the target expression can be easily sythesized. Quantitative and Qualitative results are presented on several compound expressions, and the experimental results demonstrate the feasibility and the potential of EvoGAN.

cs.CV↗

arXiv2020

The tetragonal phase of CH$_{3}$NH$_{3}$PbI$_{3}$ is strongly anharmonic

Halide perovskite (HP) semiconductors exhibit unique strong coupling between the electronic and structural dynamics. The high-temperature cubic phase of HPs is known to be entropically stabilized, with imaginary frequencies in the calculated phonon dispersion relation. Similar calculations, based on the static average crystal structure, predict a stable tetragonal phase with no imaginary modes. This work shows that in contrast to standard theory predictions, the room-temperature tetragonal phase of CH$_{3} $NH$_{3} $PbI$_{3}$ is strongly anharmonic. We use Raman polarization-orientation (PO) measurements and \textit{ab initio} molecular dynamics (AIMD) to investigate the origin and temperature evolution of the strong structural anharmonicity throughout the tetragonal phase. Raman PO measurements reveal a new spectral feature that resembles a soft mode. This mode shows an unusual continuous increase in damping with temperature which is indicative of an anharmonic potential surface. The analysis of AIMD trajectories identifies two major sources of anharmonicity: the orientational unlocking of the [CH$_{3} $NH$_{3}$]$^+$ ions and large amplitude octahedral tilting that continuously increases with temperature. Our work suggests that the standard phonon picture cannot describe the structural dynamics of tetragonal CH$_{3} $NH$_{3} $PbI$_{3}$.

cond-mat.mtrl-sci↗

arXiv2019

Halide perovskites under polarized light: Vibrational symmetry analysis using polarized Raman

In the last decade, hybrid organic-inorganic halide perovskites have emerged as a new type of semiconductor for photovoltaics and other optoelectronic applications. Unlike standard, tetrahedrally bonded semiconductors (e.g. Si and GaAs), the ionic thermal fluctuations in the halide perovskites (i.e. structural dynamics) are strongly coupled to the electronic dynamics. Therefore, it is crucial to obtain accurate and detailed knowledge about the nature of atomic motions within the crystal. This has proved to be challenging due to low thermal stability and the complex, temperature dependent structural phase sequence of the halide perovskites. Here, these challenges are overcome and a detailed analysis of the mode symmetries is provided in the low-temperature orthorhombic phase of methylammonium-lead iodide. Raman measurements using linearly- and circularly- polarized light at 1.16 eV excitation are combined with density functional perturbation theory (DFPT). By performing an iterative analysis of Raman polarization-orientation dependence and DFPT mode analysis, the crystal orientation is determined. Subsequently, accounting for birefringence effects detected using circularly polarized light excitation, the symmetries of all the observed Raman-active modes at 10 K are assigned.

cond-mat.mtrl-sci↗

arXiv2019

Quantum-critical phase out of frustrated magnetism in a strongly correlated metal

Strange-metal phenomena often develop at the border of antiferromagnetic order in strongly correlated metals. It has been well established that they can originate from the fluctuations anchored by the point of continuous quantum phase transition out of the antiferromagnetic order, i.e., a quantum critical point. What has been unclear is how these phenomena can be associated with a potential new phase of matter at zero temperature. Here we show that magnetic frustration of the 4f-local moments in the distorted Kagome intermetallic compound CePdAl gives rise to such a paramagnetic quantum-critical phase. Moreover, we demonstrate that this phase turns into a Fermi liquid through a Mott-like crossover; in a two-dimensional parameter space of pressure and magnetic field, this crossover is linked to a line of zero-temperature 4f-electron localization-delocalization phase transitions at low and moderate pressures. Our discovery motivates a new design principle for strongly correlated metallic states with unconventional excitations that may underlie the development of such effects as high temperature superconductivity.

cond-mat.str-el↗

arXiv2019

Giant magnetic quantum oscillations in the thermal conductivity of TaAs: Indications of chiral zero sound

Charge transport of topological semimetals has been in the focus of intensive investigations because of their non-trivial band topology. Heat transport of these materials, on the other hand, is largely unexplored and remains elusive. Here we report on an observation of unprecedented, giant magnetic quantum oscillations of thermal conductivity in the prototypical Weyl semimetal TaAs. The oscillations are antiphase with the quantum oscillating electronic density of states of a Weyl pocket, and their amplitudes amount to two orders of magnitude of the estimation based on the Wiedemann-Franz law. Our analyses show that all the conventional heat-transport mechanisms through diffusions of propagating electrons, phonons and electron-hole bipolar excitations, are far inadequate to account for these phenomena. Taking further experimental facts that the parallel field configuration favors much higher magneto-thermal conductivity, we propose that the newly proposed chiral zero sound provides a reasonable explanation to these exotic phenomena. More work focusing on other topological semimetals along the same line is badly called for.

cond-mat.mtrl-sci↗

arXiv2019

Finite subgroups of the extended modular group

We show that in the extended modular group PGL(2,Z) there are exactly seven finite subgroups up to conjugacy; three subgroups of size 2, one subgroup each of size 3, 4, and 6, and the trivial subgroup of size 1.

math.GR↗

arXiv2019

Superconductivity in the WP single crystal

We report the discovery of superconductivity on high-quality single crystals of transition-metal pnictides WP grown by chemical vapor transport (CVT) method. Bulk superconductivity is observed at Tc = 0.84 K under ambient pressure by electrical resistivity and AC magnetic susceptibility measurements. The effects of magnetic field on the superconducting transitions are studied, leading to a large anisotropy parameter around 2 with the in-plane and out-of-plane upper critical fields of Hc2=172 Oe and Hc2\=85 Oe, respectively. Our finding demonstrates that WP is the first superconductor in 5d transition-metal at ambient pressure in MnP-type, which will help to search for new superconductors in transition-metal pnictides.

cond-mat.supr-con↗

arXiv2018

Terahertz-Field-Induced Ferroelectricity in Quantum Paraelectric SrTiO$_3$

"Hidden phases" are metastable collective states of matter that are typically not accessible on equilibrium phase diagrams. These phases can host exotic properties in otherwise conventional materials and hence may enable novel functionality and applications, but their discovery and access are still in early stages. Using intense terahertz electric field excitation, we show that an ultrafast phase transition into a hidden ferroelectric phase can be dynamically induced in the quantum paraelectric SrTiO$_3$. The induced lowering in crystal symmetry yields dramatic changes in the phonon excitation spectra. Our results demonstrate collective coherent control over material structure, in which a single-cycle field drives ions along the microscopic pathway leading directly to their locations in a new crystalline phase on an ultrafast timescale.

cond-mat.mtrl-sci↗

arXiv2018

Generalized spatial differentiation from spin Hall effect of light

Optics naturally provides us with some powerful mathematical operations. Here we experimentally demonstrate that during reflection or refraction at a single optical planar interface, the optical computing of spatial differentiation can be realized by analyzing specific orthogonal polarization states of light. We show that the spatial differentiation is intrinsically due to the spin Hall effect of light and generally accompanies light reflection and refraction at any planar interface, regardless of material composition or incident angles. The proposed spin-optical method takes advantages of a simple and common structure to enable vectorial-field computation and perform edge detection for ultra-fast and energy-efficient image processing.

physics.optics↗

arXiv2017

Kondo destruction in a quantum paramagnet with magnetic frustration

We report results of isothermal magnetotransport and susceptibility measurements at elevated magnetic fields B down to very low temperatures T on high-quality single crystals of the frustrated Kondo-lattice system CePdAl. They reveal a B*(T) line within the paramagnetic part of the phase diagram. This line denotes a thermally broadened 'small'-to-'large' Fermi surface crossover which substantially narrows upon cooling. At B_0* = B*(T=0) = (4.6 +/- 0.1) T, this B*(T) line merges with two other crossover lines, viz. Tp(B) below and T_FL(B) above B_0*. Tp characterizes a frustration-dominated spin-liquid state, while T_FL is the Fermi-liquid temperature associated with the lattice Kondo effect. Non-Fermi-liquid phenomena which are commonly observed near a 'Kondo destruction' quantum critical point cannot be resolved in CePdAl. Our observations reveal a rare case where Kondo coupling, frustration and quantum criticality are closely intertwined.

cond-mat.str-el↗

arXiv2017

Anomalous connection between antiferromagnetic and superconducting phases in pressurized non-centrosymmetric heavy fermion compound CeRhGe3

Unconventional superconductivity frequently emerges as the transition temperature of a magnetic phase, typically antiferromagnetic, is suppressed continuously toward zero temperature. Here, we report contrary behavior in pressurized CeRhGe3, a non-centrosymmetric heavy fermion compound. We find that its pressure-tuned antiferromagnetic transition temperature (TN) appears to avoid a continuous decrease to zero temperature by terminating abruptly above a dome of pressure-induced superconductivity. Near 21.5 GPa, evidence for TN suddenly vanishes, the electrical resistance becomes linear in temperature and the superconducting transition reaches a maximum. In light of X-ray absorption spectroscopy measurements, these characteristics appear to be related to a pressured-induced Ce valence instability, which reveals as a sharp increase in the rate of change of Ce valence with applied pressure.

cond-mat.supr-con↗