arXiv subjects

Yi Li

Publications and source records attributed to Yi Li.

At least 433 records · Page 24Linked to original sources

arXiv2021

Domain Adaptation and Autoencoder Based Unsupervised Speech Enhancement

As a category of transfer learning, domain adaptation plays an important role in generalizing the model trained in one task and applying it to other similar tasks or settings. In speech enhancement, a well-trained acoustic model can be exploited to obtain the speech signal in the context of other languages, speakers, and environments. Recent domain adaptation research was developed more effectively with various neural networks and high-level abstract features. However, the related studies are more likely to transfer the well-trained model from a rich and more diverse domain to a limited and similar domain. Therefore, in this study, the domain adaptation method is proposed in unsupervised speech enhancement for the opposite circumstance that transferring to a larger and richer domain. On the one hand, the importance-weighting (IW) approach is exploited with a variance constrained autoencoder to reduce the shift of shared weights between the source and target domains. On the other hand, in order to train the classifier with the worst-case weights and minimize the risk, the minimax method is proposed. Both the proposed IW and minimax methods are evaluated from the VOICE BANK and IEEE datasets to the TIMIT dataset. The experiment results show that the proposed methods outperform the state-of-the-art approaches.

cs.SD↗

arXiv2021

Digital Twin of Electrical Tomography for Quantitative Multiphase Flow Imaging

We report a digital twin (DT) framework of electrical tomography (ET) to address the challenge of real-time quantitative multiphase flow imaging based on non-invasive and non-radioactive technologies. Multiphase flow is ubiquitous in nature, industry, and research. Accurate flow imaging is the key to understanding this complex phenomenon. Existing non-radioactive multiphase flow imaging methods based on electrical tomography are limited to providing qualitative images. The proposed DT framework, building upon a synergistic integration of 3D field coupling simulation, model-based deep learning, and edge computing, allows ET to dynamically learn the flow features in the virtual space and implement the model in the physical system, thus providing unprecedented resolution and accuracy. The DT framework is demonstrated on gas-liquid two-phase flow and electrical capacitance tomography (ECT). It can be readily extended to various tomography modalities, scenarios, and scales in biomedical, energy, and aerospace applications as an effective alternative to radioactive solutions for precise flow visualization and characterization.

physics.flu-dyn↗

arXiv2021

Local curvature estimates along the $κ$-LYZ flow

In this paper we prove a local curvature estimate for the $κ$-LYZ flow over Kähler manifolds introduced in [FGP1] and [LYZ]. In particular, we generalize the long time existence of the flow.

math.DG↗

arXiv2021

De-biased Lasso for Generalized Linear Models with A Diverging Number of Covariates

Modeling and drawing inference on the joint associations between single nucleotide polymorphisms and a disease has sparked interest in genome-wide associations studies. In the motivating Boston Lung Cancer Survival Cohort (BLCSC) data, the presence of a large number of single nucleotide polymorphisms of interest, though smaller than the sample size, challenges inference on their joint associations with the disease outcome. In similar settings, we find that neither the de-biased lasso approach (van de Geer et al. 2014), which assumes sparsity on the inverse information matrix, nor the standard maximum likelihood method can yield confidence intervals with satisfactory coverage probabilities for generalized linear models. Under this "large $n$, diverging $p$" scenario, we propose an alternative de-biased lasso approach by directly inverting the Hessian matrix without imposing the matrix sparsity assumption, which further reduces bias compared to the original de-biased lasso and ensures valid confidence intervals with nominal coverage probabilities. We establish the asymptotic distributions of any linear combinations of the parameter estimates, which lays the theoretical ground for drawing inference. Simulations show that the proposed refined de-biased estimating method performs well in removing bias and yields honest confidence interval coverage. We use the proposed method to analyze the aforementioned BLCSC data, a large scale hospital-based epidemiology cohort study, that investigates the joint effects of genetic variants on lung cancer risks.

stat.ME↗

arXiv2021

BEV-Net: Assessing Social Distancing Compliance by Joint People Localization and Geometric Reasoning

Social distancing, an essential public health measure to limit the spread of contagious diseases, has gained significant attention since the outbreak of the COVID-19 pandemic. In this work, the problem of visual social distancing compliance assessment in busy public areas, with wide field-of-view cameras, is considered. A dataset of crowd scenes with people annotations under a bird's eye view (BEV) and ground truth for metric distances is introduced, and several measures for the evaluation of social distance detection systems are proposed. A multi-branch network, BEV-Net, is proposed to localize individuals in world coordinates and identify high-risk regions where social distancing is violated. BEV-Net combines detection of head and feet locations, camera pose estimation, a differentiable homography module to map image into BEV coordinates, and geometric reasoning to produce a BEV map of the people locations in the scene. Experiments on complex crowded scenes demonstrate the power of the approach and show superior performance over baselines derived from methods in the literature. Applications of interest for public health decision makers are finally discussed. Datasets, code and pretrained models are publicly available at GitHub.

cs.CV↗

arXiv2021

Learning-Augmented Sketches for Hessians

Sketching is a dimensionality reduction technique where one compresses a matrix by linear combinations that are chosen at random. A line of work has shown how to sketch the Hessian to speed up each iteration in a second order method, but such sketches usually depend only on the matrix at hand, and in a number of cases are even oblivious to the input matrix. One could instead hope to learn a distribution on sketching matrices that is optimized for the specific distribution of input matrices. We show how to design learned sketches for the Hessian in the context of second order methods. We prove that a smaller sketching dimension of the column space of a tall matrix is possible, given an oracle that can predict the indices of the rows of large leverage score. We design such an oracle for various datasets, and this leads to a faster convergence of the well-studied iterative Hessian sketch procedure, which applies to a wide range of problems in convex optimization. We show empirically that learned sketches, compared with their "non-learned" counterparts, do improve the approximation accuracy for important problems, including LASSO and matrix estimation with nuclear norm constraints.

cs.LG↗

arXiv2021

Minimum Enclosing Parallelogram with Outliers

We study the problem of minimum enclosing rectangle with outliers, which asks to find, for a given set of $n$ planar points, a rectangle with minimum area that encloses at least $(n-t)$ points. The uncovered points are regarded as outliers. We present an exact algorithm with $O(kt^3+ktn+n^2\log n)$ runtime, assuming that no three points lie on the same line. Here $k$ denotes the number of points on the first $(t+1)$ convex layers. We further propose a sampling algorithm with runtime $O(n+\mbox{poly}(\log{n}, t, 1/ε))$, which with high probability finds a rectangle covering at least $(1-ε)(n-t)$ points with at most the exact optimal area.

cs.CG↗

arXiv2021

Expected Size of Random Tukey Layers and Convex Layers

We study the Tukey layers and convex layers of a planar point set, which consists of $n$ points independently and uniformly sampled from a convex polygon with $k$ vertices. We show that the expected number of vertices on the first $t$ Tukey layers is $O\left(kt\log(n/k)\right)$ and the expected number of vertices on the first $t$ convex layers is $O\left(kt^{3}\log(n/(kt^2))\right)$. We also show a lower bound of $Ω(t\log n)$ for both quantities in the special cases where $k=3,4$. The implications of those results in the average-case analysis of two computational geometry algorithms are then discussed.

cs.CG↗

arXiv2021

Pseudo-mask Matters in Weakly-supervised Semantic Segmentation

Most weakly supervised semantic segmentation (WSSS) methods follow the pipeline that generates pseudo-masks initially and trains the segmentation model with the pseudo-masks in fully supervised manner after. However, we find some matters related to the pseudo-masks, including high quality pseudo-masks generation from class activation maps (CAMs), and training with noisy pseudo-mask supervision. For these matters, we propose the following designs to push the performance to new state-of-art: (i) Coefficient of Variation Smoothing to smooth the CAMs adaptively; (ii) Proportional Pseudo-mask Generation to project the expanded CAMs to pseudo-mask based on a new metric indicating the importance of each class on each location, instead of the scores trained from binary classifiers. (iii) Pretended Under-Fitting strategy to suppress the influence of noise in pseudo-mask; (iv) Cyclic Pseudo-mask to boost the pseudo-masks during training of fully supervised semantic segmentation (FSSS). Experiments based on our methods achieve new state-of-art results on two changeling weakly supervised semantic segmentation datasets, pushing the mIoU to 70.0% and 40.2% on PAS-CAL VOC 2012 and MS COCO 2014 respectively. Codes including segmentation framework are released at https://github.com/Eli-YiLi/PMM

cs.CV↗

arXiv2021

Hund's coupling-assisted ferromagnetic percolation transition in a multiorbital flat band

By connecting Hund's physics with flat band physics, we establish an exact result for studying ferromagnetism in a multiorbital system. We consider a two-layer model consisting of a $p_x$, $p_y$-orbital honeycomb lattice layer and an $f$-orbital triangular lattice layer with sites aligned with the centers of the honeycomb plaquettes. The system features a flat band that admits a percolation representation for an appropriate chemical potential difference between the two layers. In this representation, the ground state space is spanned by maximum-spin clusters of localized single-particle states, and averaging over the ground states yields a correlated percolation problem with weights due to the spin degeneracy of the clusters. A paramagnetic-ferromagnetic transition occurs as the band approaches half filling and the ground states become dominated by states with a large maximum-spin cluster, as shown by Monte Carlo simulation.

cond-mat.str-el↗

arXiv2021

Template Dissolution Interfacial Patterning of Single Colloids for Nanoelectrochemistry and Nanosensing

Deterministic positioning and assembly of colloidal nanoparticles (NPs) onto substrates is a core requirement and a promising alternative to top down lithography to create functional nanostructures and nanodevices with intriguing optical, electrical, and catalytic features. Capillary-assisted particle assembly (CAPA) has emerged as an attractive technique to this end, as it allows controlled and selective assembly of a wide variety of NPs onto predefined topographical templates using capillary forces. One critical issue with CAPA, however, lies in its final printing step, where high printing yields are possible only with the use of an adhesive polymer film. To address this problem, we have developed a template dissolution interfacial patterning (TDIP) technique to assemble and print single colloidal AuNP arrays onto various dielectric and conductive substrates in the absence of any adhesion layer, with printing yields higher than 98%. The TDIP approach grants direct access to the interface between the AuNP and the target surface, enabling the use of colloidal AuNPs as building blocks for practical applications. The versatile applicability of TDIP is demonstrated by the creation of direct electrical junctions for electro- and photoelectrochemistry and nanoparticle-on-mirror geometries for single particle molecular sensing.

cond-mat.soft↗

arXiv2021

The Product of Gaussian Matrices is Close to Gaussian

We study the distribution of the {\it matrix product} $G_1 G_2 \cdots G_r$ of $r$ independent Gaussian matrices of various sizes, where $G_i$ is $d_{i-1} \times d_i$, and we denote $p = d_0$, $q = d_r$, and require $d_1 = d_{r-1}$. Here the entries in each $G_i$ are standard normal random variables with mean $0$ and variance $1$. Such products arise in the study of wireless communication, dynamical systems, and quantum transport, among other places. We show that, provided each $d_i$, $i = 1, \ldots, r$, satisfies $d_i \geq C p \cdot q$, where $C \geq C_0$ for a constant $C_0 > 0$ depending on $r$, then the matrix product $G_1 G_2 \cdots G_r$ has variation distance at most $δ$ to a $p \times q$ matrix $G$ of i.i.d.\ standard normal random variables with mean $0$ and variance $\prod_{i=1}^{r-1} d_i$. Here $δ\rightarrow 0$ as $C \rightarrow \infty$. Moreover, we show a converse for constant $r$ that if $d_i < C' \max\{p,q\}^{1/2}\min\{p,q\}^{3/2}$ for some $i$, then this total variation distance is at least $δ'$, for an absolute constant $δ' > 0$ depending on $C'$ and $r$. This converse is best possible when $p=Θ(q)$.

math.PR↗

arXiv2021

Learning to Cluster via Same-Cluster Queries

We study the problem of learning to cluster data points using an oracle which can answer same-cluster queries. Different from previous approaches, we do not assume that the total number of clusters is known at the beginning and do not require that the true clusters are consistent with a predefined objective function such as the K-means. These relaxations are critical from the practical perspective and, meanwhile, make the problem more challenging. We propose two algorithms with provable theoretical guarantees and verify their effectiveness via an extensive set of experiments on both synthetic and real-world data.

cs.LG↗

arXiv2021

Exponentially Improved Dimensionality Reduction for $\ell_1$: Subspace Embeddings and Independence Testing

Despite many applications, dimensionality reduction in the $\ell_1$-norm is much less understood than in the Euclidean norm. We give two new oblivious dimensionality reduction techniques for the $\ell_1$-norm which improve exponentially over prior ones: 1. We design a distribution over random matrices $S \in \mathbb{R}^{r \times n}$, where $r = 2^{\tilde O(d/(\varepsilon δ))}$, such that given any matrix $A \in \mathbb{R}^{n \times d}$, with probability at least $1-δ$, simultaneously for all $x$, $\|SAx\|_1 = (1 \pm \varepsilon)\|Ax\|_1$. Note that $S$ is linear, does not depend on $A$, and maps $\ell_1$ into $\ell_1$. Our distribution provides an exponential improvement on the previous best known map of Wang and Woodruff (SODA, 2019), which required $r = 2^{2^{Ω(d)}}$, even for constant $\varepsilon$ and $δ$. Our bound is optimal, up to a polynomial factor in the exponent, given a known $2^{\sqrt d}$ lower bound for constant $\varepsilon$ and $δ$. 2. We design a distribution over matrices $S \in \mathbb{R}^{k \times n}$, where $k = 2^{O(q^2)}(\varepsilon^{-1} q \log d)^{O(q)}$, such that given any $q$-mode tensor $A \in (\mathbb{R}^{d})^{\otimes q}$, one can estimate the entrywise $\ell_1$-norm $\|A\|_1$ from $S(A)$. Moreover, $S = S^1 \otimes S^2 \otimes \cdots \otimes S^q$ and so given vectors $u_1, \ldots, u_q \in \mathbb{R}^d$, one can compute $S(u_1 \otimes u_2 \otimes \cdots \otimes u_q)$ in time $2^{O(q^2)}(\varepsilon^{-1} q \log d)^{O(q)}$, which is much faster than the $d^q$ time required to form $u_1 \otimes u_2 \otimes \cdots \otimes u_q$. Our linear map gives a streaming algorithm for independence testing using space $2^{O(q^2)}(\varepsilon^{-1} q \log d)^{O(q)}$, improving the previous doubly exponential $(\varepsilon^{-1} \log d)^{q^{O(q)}}$ space bound of Braverman and Ostrovsky (STOC, 2010).

cs.DS↗

arXiv2021

Inference for High Dimensional Censored Quantile Regression

With the availability of high dimensional genetic biomarkers, it is of interest to identify heterogeneous effects of these predictors on patients' survival, along with proper statistical inference. Censored quantile regression has emerged as a powerful tool for detecting heterogeneous effects of covariates on survival outcomes. To our knowledge, there is little work available to draw inference on the effects of high dimensional predictors for censored quantile regression. This paper proposes a novel procedure to draw inference on all predictors within the framework of global censored quantile regression, which investigates covariate-response associations over an interval of quantile levels, instead of a few discrete values. The proposed estimator combines a sequence of low dimensional model estimates that are based on multi-sample splittings and variable selection. We show that, under some regularity conditions, the estimator is consistent and asymptotically follows a Gaussian process indexed by the quantile level. Simulation studies indicate that our procedure can properly quantify the uncertainty of the estimates in high dimensional settings. We apply our method to analyze the heterogeneous effects of SNPs residing in lung cancer pathways on patients' survival, using the Boston Lung Cancer Survival Cohort, a cancer epidemiology study on the molecular mechanism of lung cancer.

stat.ME↗

arXiv2021

Core shell NaErF4 at NaYF4 upconversion nanoparticles qualify a NIR speckle wavemeter by a visible CCD

Speckle patterns have been widely confirmed that can be utilized to reconstruct the wavelength information. In order to achieve higher resolution, a varies of optical diffusing waveguides have been investigated with a focus on their wavelength sensitivity. However, it has been a challenge to reach the balance among cost, volumes, resolution, and stability. In this work, we designed a compact cylindrical random scattering waveguide (CRSW) as the light diffuser only by mixing TiO2 particles and ultra-violate adhesive. The speckle patterns are generated by the light multiple scattering in the CRSW. Importantly, the thin layer of upconversion nanoparticles (UCNPs) were sprayed on the end face of the CRSW. This allows the near infrared (NIR) light to be converted to the visible light, breaking the imaging limitation of visible cameras in the NIR range. We further designed a convolution neural network (CNN) to recognize the wavelength of the speckle patterns with good robustness and excellent ability of transfer learning, resulting in the achievement of a high resolution of 20 kHz ( 0.16 fm) at around 1550 nm with temperature resistance of 2 celsius. Our results provide a low-cost, compact, and simple NIR wavemeter in particular with the ultra high resolution and good temperature stability.

physics.optics↗

arXiv2021

Single Pass Entrywise-Transformed Low Rank Approximation

In applications such as natural language processing or computer vision, one is given a large $n \times d$ matrix $A = (a_{i,j})$ and would like to compute a matrix decomposition, e.g., a low rank approximation, of a function $f(A) = (f(a_{i,j}))$ applied entrywise to $A$. A very important special case is the likelihood function $f\left( A \right ) = \log{\left( \left| a_{ij}\right| +1\right)}$. A natural way to do this would be to simply apply $f$ to each entry of $A$, and then compute the matrix decomposition, but this requires storing all of $A$ as well as multiple passes over its entries. Recent work of Liang et al.\ shows how to find a rank-$k$ factorization to $f(A)$ for an $n \times n$ matrix $A$ using only $n \cdot \operatorname{poly}(ε^{-1}k\log n)$ words of memory, with overall error $10\|f(A)-[f(A)]_k\|_F^2 + \operatorname{poly}(ε/k) \|f(A)\|_{1,2}^2$, where $[f(A)]_k$ is the best rank-$k$ approximation to $f(A)$ and $\|f(A)\|_{1,2}^2$ is the square of the sum of Euclidean lengths of rows of $f(A)$. Their algorithm uses three passes over the entries of $A$. The authors pose the open question of obtaining an algorithm with $n \cdot \operatorname{poly}(ε^{-1}k\log n)$ words of memory using only a single pass over the entries of $A$. In this paper we resolve this open question, obtaining the first single-pass algorithm for this problem and for the same class of functions $f$ studied by Liang et al. Moreover, our error is $\|f(A)-[f(A)]_k\|_F^2 + \operatorname{poly}(ε/k) \|f(A)\|_F^2$, where $\|f(A)\|_F^2$ is the sum of squares of Euclidean lengths of rows of $f(A)$. Thus our error is significantly smaller, as it removes the factor of $10$ and also $\|f(A)\|_F^2 \leq \|f(A)\|_{1,2}^2$. We also give an algorithm for regression, pointing out an error in previous work, and empirically validate our results.

cs.DS↗

arXiv2021

Improvement of image classification by multiple optical scattering

Multiple optical scattering occurs when light propagates in a non-uniform medium. During the multiple scattering, images were distorted and the spatial information they carried became scrambled. However, the image information is not lost but presents in the form of speckle patterns (SPs). In this study, we built up an optical random scattering system based on an LCD and an RGB laser source. We found that the image classification can be improved by the help of random scattering which is considered as a feedforward neural network to extracts features from image. Along with the ridge classification deployed on computer, we achieved excellent classification accuracy higher than 94%, for a variety of data sets covering medical, agricultural, environmental protection and other fields. In addition, the proposed optical scattering system has the advantages of high speed, low power consumption, and miniaturization, which is suitable for deploying in edge computing applications.

cs.CV↗