Searcharxiv⌕ Search

arXiv subjects

Tao Qian

Publications and source records attributed to Tao Qian.

At least 37 records · Page 2Linked to original sources
arXiv2022

AFD Types Sparse Representations vs. the Karhunen-Loeve Expansion for Decomposing Stochastic Processes

This article introduces adaptive Fourier decomposition (AFD) type methods, emphasizing on those that can be applied to stochastic processes and random fields, mainly including stochastic adaptive Fourier decomposition and stochastic pre-orthogonal adaptive Fourier decomposition. We establish their algorithms based on the covariant function and prove that they enjoy the same convergence rate as the Karhunen-Loève (KL) decomposition. The AFD type methods are compared with the KL decomposition. In contrast with the latter, the AFD type methods do not need to compute eigenvalues and eigenfunctions of the kernel-integral operator induced by the covariance function, and thus considerably reduce the computation complexity and computer consumes. Various kinds of dictionaries offer AFD flexibility to solve problems of a great variety, including different types of deterministic and stochastic equations. The conducted experiments show, besides the numerical convenience and fast convergence, that the AFD type decompositions outperform the KL type in describing local details, in spite of the proven global optimality of the latter.

math.ST↗
arXiv2022

SingAug: Data Augmentation for Singing Voice Synthesis with Cycle-consistent Training Strategy

Deep learning based singing voice synthesis (SVS) systems have been demonstrated to flexibly generate singing with better qualities, compared to conventional statistical parametric based methods. However, neural systems are generally data-hungry and have difficulty to reach reasonable singing quality with limited public available training data. In this work, we explore different data augmentation methods to boost the training of SVS systems, including several strategies customized to SVS based on pitch augmentation and mix-up augmentation. To further stabilize the training, we introduce the cycle-consistent training strategy. Extensive experiments on two public singing databases demonstrate that our proposed augmentation methods and the stabilizing training strategy can significantly improve the performance on both objective and subjective evaluations.

eess.AS↗
arXiv2022

Muskits: an End-to-End Music Processing Toolkit for Singing Voice Synthesis

This paper introduces a new open-source platform named Muskits for end-to-end music processing, which mainly focuses on end-to-end singing voice synthesis (E2E-SVS). Muskits supports state-of-the-art SVS models, including RNN SVS, transformer SVS, and XiaoiceSing. The design of Muskits follows the style of widely-used speech processing toolkits, ESPnet and Kaldi, for data prepossessing, training, and recipe pipelines. To the best of our knowledge, this toolkit is the first platform that allows a fair and highly-reproducible comparison between several published works in SVS. In addition, we also demonstrate several advanced usages based on the toolkit functionalities, including multilingual training and transfer learning. This paper describes the major framework of Muskits, its functionalities, and experimental results in single-singer, multi-singer, multilingual, and transfer learning scenarios. The toolkit is publicly available at https://github.com/SJTMusicTeam/Muskits.

cs.SD↗
arXiv2022

The sparse representation related with fractional heat equations

This study introduces pre-orthogonal adaptive Fourier decomposition (POAFD) to obtain approximations and numerical solutions to the fractional Laplacian initial value problem and the extension problem of Caffarelli and Silvestre (generalized Poisson equation). The method, as the first step, expands the initial data function into a sparse series of the fundamental solutions with fast convergence, and, as the second step, makes use the semigroup or the reproducing kernel property of each of the expanding entries. Experiments show effectiveness and efficiency of the proposed series solutions.

math.NA↗
arXiv2022

Hilbert Boundary Value Problems for Hyper Monogenic Functions on The Hyperplane

This paper systematically studies Hilbert boundary value problems for hyper monogenic functions on the hyperplane for the solutions being of any integer orders at the infinity, where the negative order cases are new even when restricted to the complex plane context. The explicit solution formulas are given and the solvability conditions are specified. The results are proved through using the Clifford symmetric extension method to reduce Hilbert boundary value problems to Riemann boundary value problems.

math.CV↗
arXiv2022

Sparse Representations of Solutions to a class of Random Boundary Value Problems

We introduce certain sparse representation methods, named as stochastic pre-orthogonal adaptive Fourier decomposition 1 and 2 (SPOAFD1 and SPOAFD2) to solve the Dirichlet boundary value problem and the Cauchy initial value problem of random data. To solve the stochastic boundary value problems the sparse representation is, as the initial step, applied to the random boundary data. Due to the semigroup property of the Poisson and the heat kernel, each entry of the expanding series can be lifted up to compose a solution of the Dirichlet and the Cauchy initial value problem, respectively. The sparse representation gives rise to analytic as well as numerical solutions to the problems with high efficiency.

math.NA↗
arXiv2022

n-Best Kernel Approximation in Reproducing Kernel Hilbert Spaces

By making a seminal use of the maximum modulus principle of holomorphic functions we prove existence of $n$-best kernel approximation for a wide class of reproducing kernel Hilbert spaces of holomorphic functions in the unit disc, and for the corresponding class of Bochner type spaces of stochastic processes. This study thus generalizes the classical result of $n$-best rational approximation for the Hardy space and a recent result of $n$-best kernel approximation for the weighted Bergman spaces of the unit disc. The type of approximations have significant applications to signal and image processing and system identification, as well as to numerical solutions of the classical and the stochastic type integral and differential equations.

math.CV↗
arXiv2021

The Poisson kernel and the Fourier transform of the slice monogenic Cauchy kernels

The Fueter-Sce-Qian (FSQ for short) mapping theorem is a two-steps procedure to extend holomorphic functions of one complex variable to slice monogenic functions and to monogenic functions. Using the Cauchy formula of slice monogenic functions the FSQ-theorem admits an integral representation for $n$ odd. In this paper we show that the relation $ Δ_{n+1}^{(n-1)/2}S_L^{-1}=\mathcal{F}^L_n $ between the slice monogenic Cauchy kernel $S_L^{-1}$ and the F-kernel $\mathcal{F}^L_n$, that appear in the integral form of the FSQ-theorem for $n$ odd, holds also in the case we consider the fractional powers of the Laplace operator $Δ_{n+1}$ in dimension $n+1$, i.e., for $n$ even. Moreover, this relation is proven computing explicitly Fourier transform of the kernels $S_L^{-1}$ and $\mathcal{F}^L_n$ as functions of the Poisson kernel. Similar results hold for the right kernels $S_R^{-1}$ and of $\mathcal{F}^R_n$.

math.CV↗
arXiv2021

Adaptive Fourier decomposition of slice regular functions

In the slice Hardy space over the unit ball of quaternions, we introduce the slice hyperbolic backward shift operator $\mathcal S_a$ with the decomposition process $$f=e_a\langle f, e_a\rangle+B_{a}*\mathcal S_a f,$$ where $e_a$ denotes the slice normalized Szegö kernel and $ B_a $ the slice Blaschke factor. Iterating the above decomposition process, a corresponding maximal selection principle gives rise to the slice adaptive Fourier decomposition. This leads to a adaptive slice Takenaka-Malmquist orthonormal system.

math.CV↗
arXiv2021

On Monogenic Reproducing Kernel Hilbert Spaces of the Paley-Wiener Type

In the Clifford algebra setting the present study develops three reproducing kernel Hilbert spaces of the Paley-Wiener type, namely the Paley-Wiener spaces, the Hardy spaces on strips, and the Bergman spaces on strips. In particular, we give spectrum characterizations and representation formulas of the functions in those spaces and estimation of their respective reproducing kernels.

math.CV↗
arXiv2021

A Granular Sieving Algorithm for Deterministic Global Optimization

A gradient-free deterministic method is developed to solve global optimization problems for Lipschitz continuous functions defined in arbitrary path-wise connected compact sets in Euclidean spaces. The method can be regarded as granular sieving with synchronous analysis in both the domain and range of the objective function. With straightforward mathematical formulation applicable to both univariate and multivariate objective functions, the global minimum value and all the global minimizers are located through two decreasing sequences of compact sets in, respectively, the domain and range spaces. The algorithm is easy to implement with moderate computational cost. The method is tested against extensive benchmark functions in the literature. The experimental results show remarkable effectiveness and applicability of the algorithm.

math.OC↗
arXiv2021

Gravitational Wave From Axion-like Particle Inflation

In this paper, we investigate the Axion-like Particle inflation by applying the multi-nature inflation model, where the end of inflation is achieved through the phase transition (PT). The events of PT should not be less than $200$, which results in the free parameter $n\geq404$. Under the latest CMB restrictions, we found that the inflation energy is fixed at $10^{15} \rm{GeV}$. Then, we deeply discussed the corresponding stochastic background of the primordial gravitational wave (GW) during inflation. We study the two kinds of $n$ cases, i.e., $n=404, 2000$. We observe that the magnitude of $n$ is negligible for the physical observations, such as $n_s$, $r$, $Λ$, and $Ω_{\rm{GW}}h^2$. In the low-frequency regions, the GW is dominated by the quantum fluctuations, and this GW can be detected by Decigo at $10^{-1}~\rm{Hz}$. However, GW generated by PT dominates the high-frequency regions, which is expected to be detected by future 3DSR detector.

hep-ph↗
arXiv2021

Orthogonalization in Clifford Hilbert modules and applications

We prove that the Gram--Schmidt orthogonalization process can be carried out in Hilbert modules over Clifford algebras, in spite of the un-invertibility and the un-commutativity of general Clifford numbers. Then we give two crucial applications of the orthogonalization method. One is to give a constructive proof of existence of an orthonormal basis of the inner spherical monogenics of order $k$ for each $k\in\mathbb{N}.$ The second is to formulate the Clifford Takenaka--Malmquist systems, or in other words, the Clifford rational orthogonal systems, as well as define Clifford Blaschke product functions, in both the unit ball and the half space contexts. The Clifford TM systems then are further used to establish an adaptive rational approximation theory for $L^2$ functions on the sphere and in $\mathbb{R}^m.$

math.FA↗
arXiv2020

A Sparse Representation of Random Signals

Studies of sparse representation of deterministic signals have been well developed. Amongst there exists one called adaptive Fourier decomposition (AFD) established through adaptive selections of the parameters defining a Takenaka-Malmquist system in one-complex variable. The AFD type algorithms give rise to sparse representations of signals of finite energy. The multivariate generalization of AFD is one called pre-orthogonal AFD (POAFD), the latter being established with the context Hilbert space possessing a dictionary. The purpose of the present study is to generalize both AFD and POAFD to random signals. We work on two types of random signals. One is those expressible as the sum of a deterministic signal with an error term such as a white noise; and the other is, in general, as mixture of several classes of random signals obeying certain distributive law. In the first part of the paper we develop an AFD type sparse representation for one-dimensional random signals by making use analysis of one complex variable. In the second part, without complex analysis, we treat multivariate random signals in the context of stochastic Hilbert space with a dictionary. Like in the deterministic signal case the established random sparse representations are powerful tools in practical signal analysis.

math.PR↗
arXiv2020

Reproducing kernels of some weighted Bergman spaces

Herein, the theory of Bergman kernel is developed to the weighted case. A general form of weighted Bergman reproducing kernel is obtained, by which we can calculate concrete Bergman kernel functions for specific weights and domains.

math.CV↗
arXiv2020

Sparse Approximation to the Dirac-δ Distribution

The Dirac-δ distribution may be realized through sequences of convlutions, the latter being also regarded as approximation to the identity. The present study proposes the so called pre-orthogonal adaptive Fourier decomposition (POAFD) method to realize fast approximation to the identity. The type of sparse representation method has potential applications in signal and image analysis, as well as in system identification.

math.CA↗