Searcharxiv⌕ Search

arXiv subjects

Yan Yang

Publications and source records attributed to Yan Yang.

At least 127 records · Page 7Linked to original sources
arXiv2022

Turbulence compressibility reduction with helicity

Numerical test of isotropic turbulence compressibility reduction with helicity in a cyclic box is performed. The ratios of compressibility-relevant-mode spectra over those of kinetic energy present power laws at large wavenumbers in the dissipation range, indicating a common difference of $11/15$ in the exponents of the algebraic prefactor of the nonhelical power spectra over those of helical ones. Our results being not derived from the shapes of the spectra themselves, the implied information about the helicity effect on the complex singularities of the discretized dynamical system can still be of reasonable value for insights of the Navier-Stokes equation, although the high-order finite difference scheme used for computation may not be as accurate in dissipation range as the state-of-the-art of incompressible turbulence with pseudo-spectral method. Possible applications in controlling flows, for the purposes of, say, decreasing turbulence noise, are also discussed according to the spectral fluctuations.

physics.flu-dyn↗
arXiv2022

Exemplar Guided Deep Neural Network for Spatial Transcriptomics Analysis of Gene Expression Prediction

Spatial transcriptomics (ST) is essential for understanding diseases and developing novel treatments. It measures gene expression of each fine-grained area (i.e., different windows) in the tissue slide with low throughput. This paper proposes an Exemplar Guided Network (EGN) to accurately and efficiently predict gene expression directly from each window of a tissue slide image. We apply exemplar learning to dynamically boost gene expression prediction from nearest/similar exemplars of a given tissue slide image window. Our EGN framework composes of three main components: 1) an extractor to structure a representation space for unsupervised exemplar retrievals; 2) a vision transformer (ViT) backbone to progressively extract representations of the input window; and 3) an Exemplar Bridging (EB) block to adaptively revise the intermediate ViT representations by using the nearest exemplars. Finally, we complete the gene expression prediction task with a simple attention-based prediction block. Experiments on standard benchmark datasets indicate the superiority of our approach when comparing with the past state-of-the-art (SOTA) methods.

cs.CV↗
arXiv2022

ISG: I can See Your Gene Expression

This paper aims to predict gene expression from a histology slide image precisely. Such a slide image has a large resolution and sparsely distributed textures. These obstruct extracting and interpreting discriminative features from the slide image for diverse gene types prediction. Existing gene expression methods mainly use general components to filter textureless regions, extract features, and aggregate features uniformly across regions. However, they ignore gaps and interactions between different image regions and are therefore inferior in the gene expression task. Instead, we present ISG framework that harnesses interactions among discriminative features from texture-abundant regions by three new modules: 1) a Shannon Selection module, based on the Shannon information content and Solomonoff's theory, to filter out textureless image regions; 2) a Feature Extraction network to extract expressive low-dimensional feature representations for efficient region interactions among a high-resolution image; 3) a Dual Attention network attends to regions with desired gene expression features and aggregates them for the prediction task. Extensive experiments on standard benchmark datasets show that the proposed ISG framework outperforms state-of-the-art methods significantly.

cs.CV↗
arXiv2022

Accelerating Time-Reversal Imaging with Neural Operators for Real-time Earthquake Locations

Earthquake hypocenters form the basis for a wide array of seismological analyses. Pick-based earthquake location workflows rely on the accuracy of phase pickers and may be biased when dealing with complex earthquake sequences in heterogeneous media. Time-reversal imaging of passive seismic sources with the cross-correlation imaging condition has potential for earthquake location with high accuracy and high resolution, but carries a large computational cost. Here we present an alternative deep-learning approach for earthquake location by combining the benefits of neural operators for wave propagation and time reversal imaging with multi-station waveform recordings. A U-shaped neural operator is trained to propagate seismic waves with various source time functions and thus can predict a backpropagated wavefield for each station in negligible time. These wavefields can either be stacked or correlated to locate earthquakes from the resulting source images. Compared with other waveform-based deep-learning location methods, time reversal imaging accounts for physical laws of wave propagation and is expected to achieve accurate earthquake location. We demonstrate the method with the 2D acoustic wave equation on both synthetic and field data. The results show that our method can efficiently obtain high resolution and high accuracy correlation-based time reversal imaging of earthquake sources. Moreover, our approach is adaptable to the number and geometry of seismic stations, which opens new strategies for real-time earthquake location and monitoring with dense seismic networks.

physics.geo-ph↗
arXiv2022

Less is More: Facial Landmarks can Recognize a Spontaneous Smile

Smile veracity classification is a task of interpreting social interactions. Broadly, it distinguishes between spontaneous and posed smiles. Previous approaches used hand-engineered features from facial landmarks or considered raw smile videos in an end-to-end manner to perform smile classification tasks. Feature-based methods require intervention from human experts on feature engineering and heavy pre-processing steps. On the contrary, raw smile video inputs fed into end-to-end models bring more automation to the process with the cost of considering many redundant facial features (beyond landmark locations) that are mainly irrelevant to smile veracity classification. It remains unclear to establish discriminative features from landmarks in an end-to-end manner. We present a MeshSmileNet framework, a transformer architecture, to address the above limitations. To eliminate redundant facial features, our landmarks input is extracted from Attention Mesh, a pre-trained landmark detector. Again, to discover discriminative features, we consider the relativity and trajectory of the landmarks. For the relativity, we aggregate facial landmark that conceptually formats a curve at each frame to establish local spatial features. For the trajectory, we estimate the movements of landmark composed features across time by self-attention mechanism, which captures pairwise dependency on the trajectory of the same landmark. This idea allows us to achieve state-of-the-art performances on UVA-NEMO, BBC, MMI Facial Expression, and SPOS datasets.

cs.CV↗
arXiv2022

Physical interpretation of nonlocal quantum correlation through local description of subsystems

Characterization and categorization of quantum correlations are both fundamentally and practically important in quantum information science. Although quantum correlations such as non-separability, steerability, and non-locality can be characterized by different theoretical models in different scenarios with either known (trusted) or unknown (untrusted) knowledge of the associated systems, such characterization sometimes lacks unambiguous to experimentalist. In this work, we propose the physical interpretation of nonlocal quantum correlation between two systems. In the absence of {\it complete local description} of one of the subsystems quantified by the {\it local uncertainty relation}, the correlation between subsystems becomes nonlocal. Remarkably, different nonlocal quantum correlations can be discriminated from a single uncertainty relation derived under local hidden state (LHS)-LHS model only. We experimentally characterize the two-qubit Werner state in different scenarios.

quant-ph↗
arXiv2022

Strategies for determining the cascade rate in MHD turbulence: isotropy, anisotropy, and spacecraft sampling

``Exact'' laws for evaluating cascade rates, tracing back to the Kolmogorov ``4/5'' law, have been extended to many systems of interest including magnetohydrodynamics (MHD), and compressible flows of the magnetofluid and ordinary fluid types. It is understood that implementations may be limited by the quantity of available data and by the lack of turbulence symmetry. Assessment of the accuracy and feasibility of such ``third-order'' (or Yaglom) relations is most effectively accomplished by examining the von Karman-Howarth equation in increment form, a framework from which the third-order laws are derived as asymptotic approximations. Using this approach, we examine the context of third-order laws for incompressible MHD in some detail. The simplest versions rely on the assumption of isotropy and the presence of a well-defined inertial range, while related procedures generalize the same idea to arbitrary rotational symmetries. Conditions for obtaining correct and accurate values of the dissipation rate from these laws based on several sampling and fitting strategies are investigated using results from simulations. The questions we address are of particular relevance to sampling of solar wind turbulence by one or more spacecraft.

physics.space-ph↗
arXiv2022

Hardness prediction of age-hardening aluminum alloy based on ensemble learning

With the rapid development of artificial intelligence, the combination of material database and machine learning has driven the progress of material informatics. Because aluminum alloy is widely used in many fields, so it is significant to predict the properties of aluminum alloy. In this thesis, the data of Al-Cu-Mg-X (X: Zn, Zr, etc.) alloy are used to input the composition, aging conditions (time and temperature) and predict its hardness. An ensemble learning solution based on automatic machine learning and an attention mechanism introduced into the secondary learner of deep neural network are proposed respectively. The experimental results show that selecting the correct secondary learner can further improve the prediction accuracy of the model. This manuscript introduces the attention mechanism to improve the secondary learner based on deep neural network, and obtains a fusion model with better performance. The R-Square of the best model is 0.9697 and the MAE is 3.4518HV.

cond-mat.mtrl-sci↗
arXiv2022

K-textures, a self-supervised hard clustering deep learning algorithm for satellite image segmentation

Deep learning self-supervised algorithms that can segment an image in a fixed number of hard labels such as the k-means algorithm and relying only on deep learning techniques are still lacking. Here, we introduce the k-textures algorithm which provides self-supervised segmentation of a 4-band image (RGB-NIR) for a $k$ number of classes. An example of its application on high resolution Planet satellite imagery is given. Our algorithm shows that discrete search is feasible using convolutional neural networks (CNN) and gradient descent. The model detects $k$ hard clustering classes represented in the model as $k$ discrete binary masks and their associated $k$ independently generated textures, that combined are a simulation of the original image. The similarity loss is the mean squared error between the features of the original and the simulated image, both extracted from the penultimate convolutional block of Keras 'imagenet' pretrained VGG-16 model and a custom feature extractor made with Planet data. The main advances of the k-textures model are: first, the $k$ discrete binary masks are obtained inside the model using gradient descent. The model allows for the generation of discrete binary masks using a novel method using a hard sigmoid activation function. Second, it provides hard clustering classes -- each pixels has only one class. Finally, in comparison to k-means, where each pixel is considered independently, here, contextual information is also considered and each class is not associated only to similar values in the color channels but also to a texture. Our approach is designed to ease the production of training samples for satellite image segmentation and the k-textures architecture could be adapted to support different number of bands and for more complex tasks, such as object self-segmentation. The model codes and weights are available at https://doi.org/10.5281/zenodo.6359859

cs.CV↗
arXiv2022

DPST: De Novo Peptide Sequencing with Amino-Acid-Aware Transformers

De novo peptide sequencing aims to recover amino acid sequences of a peptide from tandem mass spectrometry (MS) data. Existing approaches for de novo analysis enumerate MS evidence for all amino acid classes during inference. It leads to over-trimming on receptive fields of MS data and restricts MS evidence associated with following undecoded amino acids. Our approach, DPST, circumvents these limitations with two key components: (1) A confidence value aggregation encoder to sketch spectrum representations according to amino-acid-based connectivity among MS; (2) A global-local fusion decoder to progressively assimilate contextualized spectrum representations with a predefined preconception of localized MS evidence and amino acid priors. Our components originate from a closed-form solution and selectively attend to informative amino-acid-aware MS representations. Through extensive empirical studies, we demonstrate the superiority of DPST, showing that it outperforms state-of-the-art approaches by a margin of 12% - 19% peptide accuracy.

q-bio.QM↗
arXiv2022

Pressure-Strain Interaction as the Energy Dissipation Estimate in Collisionless Plasma

The dissipative mechanism in weakly collisional plasma is a topic that pervades decades of studies without a consensus solution. We compare several energy dissipation estimates based on energy transfer processes in plasma turbulence and provide justification for the pressure-strain interaction as a direct estimate of the energy dissipation rate. The global and scale-by-scale energy balances are examined in 2.5D and 3D kinetic simulations. We show that the global internal energy increase and the temperature enhancement of each species are directly tracked by the pressure-strain interaction. The incompressive part of the pressure-strain interaction dominates over its compressive part in all simulations considered. The scale-by-scale energy balance is quantified by scale filtered Vlasov-Maxwell equations, a kinetic plasma approach, and the lag dependent von Kármán-Howarth equation, an approach based on fluid models. We find that the energy balance is exactly satisfied across all scales, but the lack of a well-defined inertial range influences the distribution of the energy budget among different terms in the inertial range. Therefore, the widespread use of the Yaglom relation to estimating dissipation rate is questionable in some cases, especially when the scale separation in the system is not clearly defined. In contrast, the pressure-strain interaction balances exactly the dissipation rate at kinetic scales regardless of the scale separation.

physics.space-ph↗
arXiv2022

Molecular distance matrix prediction based on graph convolutional networks

Molecular structure has important applications in many fields. For example, some studies show that molecular spatial information can be used to achieve better prediction results when predicting molecular properties. However, traditional molecular geometry calculations, such as density functional theory (DFT), are time-consuming. In view of this, we propose a model based on graph convolutional networks to predict the pairwise distance between atoms, also called distance matrix prediction of the molecule(DMGCN). In order to indicate the effect of DMGCN model, the model is compared with the model DeeperGCN-DAGNN and the method of calculating molecular conformation in RDKit. Results show that the MAE of DMGCN is smaller than DeeperGCN-DAGNN and RDKit. In addition, the distances predicted by the DMGCN model and the distances calculated by the QM9 dataset are used to predict the molecular properties, thus showing the effectiveness of the distance predicted by the DMGCN model.

physics.chem-ph↗
arXiv2021

Effects of forcing mechanisms on the multiscale properties of magnetohydrodynamics

We performed numerical simulations to study the response of magnetohydrodynamics (MHD) to large-scale stochastic forcing mechanisms parametrized by one parameter, $0 \le a \le1$, going from direct injection on the velocity field ($a = 1$) to stirring acts on the magnetic field only ($a = 0$). We study the multi-scale properties of the energy transfer, by splitting the total flux in channels mediated by (i) the kinetic non-linear advection, (ii) the Lorentz force, (iii) the magnetic advection and (iv) magnetic stretching term. We further decompose the fluxes in two sub-channels given by heterochiral and homochiral components in order to distinguish forward, inverse and flux-loop cascades. We show that there exists a quasi-singular role of the magnetic forcing mechanism for $a \sim 1$: a small injection on the magnetic field $a < 1$ can strongly deplete the mean flux of kinetic energy transfer throughout the kinetic non-linear advection channel. We also show that this negligible mean flux is the result of a flux-loop balance between heterochiral (direct) and homochiral (inverse) transfers. Conversely, both homochiral and heterochiral channels transfer energy forward for the other three channels. Cross exchange between velocity and the magnetic field is reversed around $a = 0.4$ and except when $a \sim 1$ we always observe that heterochiral mixed velocity-magnetic energy triads tend to move energy from magnetic to velocity fields. Our study is an attempt to further characterize the multi-scale nature of MHD dynamics, by disentangling different properties of the total energy transfer mechanisms, which can be useful for improving sub-grid-modelling.

physics.plasm-ph↗
arXiv2021

Feature Modulation to Improve Struggle Detection in Web Search: A Psychological Approach

Searcher struggle is important feedback to Web search engines. Existing Web search struggle detection methods rely on effort-based features to identify the struggling moments. Their underlying assumption is that the more effort a user spends, the more struggling the user may be. However, recent studies have suggested this simple association might be incorrect. This paper proposes a new feature modulation method for struggle detection and refers to the reversal theory in psychology. The reversal theory (RT) points out that instead of having a static personality trait, people constantly switch between opposite psychological states, complicating the relationship between the efforts they spend and the level of frustration they feel. Supported by the theory, our method modulates the effort-based features based on RT's bi-modal arousal model. Evaluations on week-long Web search logs confirm that the proposed method can statistically significantly improve state-of-the-art struggle detection methods.

cs.IR↗
arXiv2021

DP color functions versus chromatic polynomials

For any graph $G$, the chromatic polynomial of $G$ is the function $P(G,m)$ which counts the number of proper $m$-colorings of $G$ for each positive integer $m$. The DP color function $P_{DP}(G,m)$ of $G$, introduced by Kaul and Mudrock in 2019, is a generalization of $P(G,m)$ with $P_{DP}(G,m)\le P(G,m)$ for each positive integer $m$. Let $P_{DP}(G)\approx P(G)$ (resp. $P_{DP}(G)< P(G)$) denote the property that $P_{DP}(G,m)=P(G,m)$ (resp. $P_{DP}(G,m)<P(G,m)$) holds for sufficiently large integers $m$.It is an interesting problem of finding graphs $G$ for which $P_{DP}(G)\approx P(G)$ (resp. $P_{DP}(G,m)<P(G,m)$) holds. Kaul and Mudrock showed that if $G$ has an even girth, then $P_{DP}(G)<P(G)$ and Mudrock and Thomason recently proved that $P_{DP}(G)\approx P(G)$ holds for each graph $G$ which has a dominating vertex. We shall generalize their results in this article. For each edge $e$ in $G$, let $\ell(e)=\infty$ if $e$ is a bridge of $G$, and let $\ell(e)$ be the length of a shortest cycle in $G$ containing $e$ otherwise. We first show that if $\ell(e)$ is even for some edge $e$ in $G$, then $P_{DP}(G)<P(G)$ holds. However, the converse statement of this conclusion fails with infinitely many counterexamples. We then prove that $P_{DP}(G)\approx P(G)$ holds for every graph $G$ that contains a spanning tree $T$ such that for each $e\in E(G)\setminus E(T)$, $\ell(e)$ is odd and $e$ contained in a cycle $C$ of length $\ell (e)$ with the property that $\ell(e')<\ell(e)$ for each $e'\in E(C)\setminus (E(T)\cup \{e\})$. Some open problems are proposed in this article.

math.CO↗
arXiv2021

Seismic wave propagation and inversion with Neural Operators

Seismic wave propagation forms the basis for most aspects of seismological research, yet solving the wave equation is a major computational burden that inhibits the progress of research. This is exacerbated by the fact that new simulations must be performed when the velocity structure or source location is perturbed. Here, we explore a prototype framework for learning general solutions using a recently developed machine learning paradigm called Neural Operator. A trained Neural Operator can compute a solution in negligible time for any velocity structure or source location. We develop a scheme to train Neural Operators on an ensemble of simulations performed with random velocity models and source locations. As Neural Operators are grid-free, it is possible to evaluate solutions on higher resolution velocity models than trained on, providing additional computational efficiency. We illustrate the method with the 2D acoustic wave equation and demonstrate the method's applicability to seismic tomography, using reverse mode automatic differentiation to compute gradients of the wavefield with respect to the velocity structure. The developed procedure is nearly an order of magnitude faster than using conventional numerical methods for full waveform inversion.

physics.geo-ph↗
arXiv2021

S2FGAN: Semantically Aware Interactive Sketch-to-Face Translation

Interactive facial image manipulation attempts to edit single and multiple face attributes using a photo-realistic face and/or semantic mask as input. In the absence of the photo-realistic image (only sketch/mask available), previous methods only retrieve the original face but ignore the potential of aiding model controllability and diversity in the translation process. This paper proposes a sketch-to-image generation framework called S2FGAN, aiming to improve users' ability to interpret and flexibility of face attribute editing from a simple sketch. The proposed framework modifies the constrained latent space semantics trained on Generative Adversarial Networks (GANs). We employ two latent spaces to control the face appearance and adjust the desired attributes of the generated face. Instead of constraining the translation process by using a reference image, the users can command the model to retouch the generated images by involving the semantic information in the generation process. In this way, our method can manipulate single or multiple face attributes by only specifying attributes to be changed. Extensive experimental results on CelebAMask-HQ dataset empirically shows our superior performance and effectiveness on this task. Our method successfully outperforms state-of-the-art methods on attribute manipulation by exploiting greater control of attribute intensity.

cs.CV↗