arXiv subjects

Xin Qi

Publications and source records attributed to Xin Qi.

At least 37 records · Page 2Linked to original sources

arXiv2024

PPPR: Portable Plug-in Prompt Refiner for Text to Audio Generation

Text-to-Audio (TTA) aims to generate audio that corresponds to the given text description, playing a crucial role in media production. The text descriptions in TTA datasets lack rich variations and diversity, resulting in a drop in TTA model performance when faced with complex text. To address this issue, we propose a method called Portable Plug-in Prompt Refiner, which utilizes rich knowledge about textual descriptions inherent in large language models to effectively enhance the robustness of TTA acoustic models without altering the acoustic training set. Furthermore, a Chain-of-Thought that mimics human verification is introduced to enhance the accuracy of audio descriptions, thereby improving the accuracy of generated content in practical applications. The experiments show that our method achieves a state-of-the-art Inception Score (IS) of 8.72, surpassing AudioGen, AudioLDM and Tango.

cs.SD↗

arXiv2024

Generalized Fake Audio Detection via Deep Stable Learning

Although current fake audio detection approaches have achieved remarkable success on specific datasets, they often fail when evaluated with datasets from different distributions. Previous studies typically address distribution shift by focusing on using extra data or applying extra loss restrictions during training. However, these methods either require a substantial amount of data or complicate the training process. In this work, we propose a stable learning-based training scheme that involves a Sample Weight Learning (SWL) module, addressing distribution shift by decorrelating all selected features via learning weights from training samples. The proposed portable plug-in-like SWL is easy to apply to multiple base models and generalizes them without using extra data during training. Experiments conducted on the ASVspoof datasets clearly demonstrate the effectiveness of SWL in generalizing different models across three evaluation datasets from different distributions.

cs.SD↗

arXiv2024

Relation Modeling and Distillation for Learning with Noisy Labels

Learning with noisy labels has become an effective strategy for enhancing the robustness of models, which enables models to better tolerate inaccurate data. Existing methods either focus on optimizing the loss function to mitigate the interference from noise, or design procedures to detect potential noise and correct errors. However, their effectiveness is often compromised in representation learning due to the dilemma where models overfit to noisy labels. To address this issue, this paper proposes a relation modeling and distillation framework that models inter-sample relationships via self-supervised learning and employs knowledge distillation to enhance understanding of latent associations, which mitigate the impact of noisy labels. Specifically, the proposed method, termed RMDNet, includes two main modules, where the relation modeling (RM) module implements the contrastive learning technique to learn representations of all data, an unsupervised approach that effectively eliminates the interference of noisy tags on feature extraction. The relation-guided representation learning (RGRL) module utilizes inter-sample relation learned from the RM module to calibrate the representation distribution for noisy samples, which is capable of improving the generalization of the model in the inference phase. Notably, the proposed RMDNet is a plug-and-play framework that can integrate multiple methods to its advantage. Extensive experiments were conducted on two datasets, including performance comparison, ablation study, in-depth analysis and case study. The results show that RMDNet can learn discriminative representations for noisy data, which results in superior performance than the existing methods.

cs.AI↗

arXiv2024

Robust Visual Tracking via Iterative Gradient Descent and Threshold Selection

Visual tracking fundamentally involves regressing the state of the target in each frame of a video. Despite significant progress, existing regression-based trackers still tend to experience failures and inaccuracies. To enhance the precision of target estimation, this paper proposes a tracking technique based on robust regression. Firstly, we introduce a novel robust linear regression estimator, which achieves favorable performance when the error vector follows i.i.d Gaussian-Laplacian distribution. Secondly, we design an iterative process to quickly solve the problem of outliers. In fact, the coefficients are obtained by Iterative Gradient Descent and Threshold Selection algorithm (IGDTS). In addition, we expend IGDTS to a generative tracker, and apply IGDTS-distance to measure the deviation between the sample and the model. Finally, we propose an update scheme to capture the appearance changes of the tracked object and ensure that the model is updated correctly. Experimental results on several challenging image sequences show that the proposed tracker outperformance existing trackers.

cs.CV↗

arXiv2024

Comparative Study of Neighbor-based Methods for Local Outlier Detection

The neighbor-based method has become a powerful tool to handle the outlier detection problem, which aims to infer the abnormal degree of the sample based on the compactness of the sample and its neighbors. However, the existing methods commonly focus on designing different processes to locate outliers in the dataset, while the contributions of different types neighbors to outlier detection has not been well discussed. To this end, this paper studies the neighbor in the existing outlier detection algorithms and a taxonomy is introduced, which uses the three-level components of information, neighbor and methodology to define hybrid methods. This taxonomy can serve as a paradigm where a novel neighbor-based outlier detection method can be proposed by combining different components in this taxonomy. A large number of comparative experiments were conducted on synthetic and real-world datasets in terms of performance comparison and case study, and the results show that reverse K-nearest neighbor based methods achieve promising performance and dynamic selection method is suitable for working in high-dimensional space. Notably, it is verified that rationally selecting components from this taxonomy may create an algorithms superior to existing methods.

cs.LG↗

arXiv2023

Trend-Based SAC Beam Control Method with Zero-Shot in Superconducting Linear Accelerator

The superconducting linear accelerator is a highly flexiable facility for modern scientific discoveries, necessitating weekly reconfiguration and tuning. Accordingly, minimizing setup time proves essential in affording users with ample experimental time. We propose a trend-based soft actor-critic(TBSAC) beam control method with strong robustness, allowing the agents to be trained in a simulated environment and applied to the real accelerator directly with zero-shot. To validate the effectiveness of our method, two different typical beam control tasks were performed on China Accelerator Facility for Superheavy Elements (CAFe II) and a light particle injector(LPI) respectively. The orbit correction tasks were performed in three cryomodules in CAFe II seperately, the time required for tuning has been reduced to one-tenth of that needed by human experts, and the RMS values of the corrected orbit were all less than 1mm. The other transmission efficiency optimization task was conducted in the LPI, our agent successfully optimized the transmission efficiency of radio-frequency quadrupole(RFQ) to over $85\%$ within 2 minutes. The outcomes of these two experiments offer substantiation that our proposed TBSAC approach can efficiently and effectively accomplish beam commissioning tasks while upholding the same standard as skilled human experts. As such, our method exhibits potential for future applications in other accelerator commissioning fields.

physics.acc-ph↗

arXiv2023

Transport of intense ion beams in plasmas: collimation and energy-loss reduction

We compare the transport properties of a well-characterized hydrogen plasma for low and high current ion beams. The energy-loss of low current beams can be well understood, within the framework of current stopping power models. However, for high current proton beams, significant energy-loss reduction and collimation is observed in the experiment. We have developed a new particle-in-cell code, which includes both collective electromagnetic effects and collisional interactions. Our simulations indicate that resistive magnetic fields, induced by the transport of an intense proton beam, act to collimate the proton beam and simultaneously deplete the local plasma density along the beam path. This in turn causes the energy-loss reduction detected in the experiment.

physics.plasm-ph↗

arXiv2022

Cross-Modal ASR Post-Processing System for Error Correction and Utterance Rejection

Although modern automatic speech recognition (ASR) systems can achieve high performance, they may produce errors that weaken readers' experience and do harm to downstream tasks. To improve the accuracy and reliability of ASR hypotheses, we propose a cross-modal post-processing system for speech recognizers, which 1) fuses acoustic features and textual features from different modalities, 2) joints a confidence estimator and an error corrector in multi-task learning fashion and 3) unifies error correction and utterance rejection modules. Compared with single-modal or single-task models, our proposed system is proved to be more effective and efficient. Experiment result shows that our post-processing system leads to more than 10% relative reduction of character error rate (CER) for both single-speaker and multi-speaker speech on our industrial ASR system, with about 1.7ms latency for each token, which ensures that extra latency introduced by post-processing is acceptable in streaming speech recognition.

eess.AS↗

arXiv2020

AE-OT-GAN: Training GANs from data specific latent distribution

Though generative adversarial networks (GANs) areprominent models to generate realistic and crisp images,they often encounter the mode collapse problems and arehard to train, which comes from approximating the intrinsicdiscontinuous distribution transform map with continuousDNNs. The recently proposed AE-OT model addresses thisproblem by explicitly computing the discontinuous distribu-tion transform map through solving a semi-discrete optimaltransport (OT) map in the latent space of the autoencoder.However the generated images are blurry. In this paper, wepropose the AE-OT-GAN model to utilize the advantages ofthe both models: generate high quality images and at thesame time overcome the mode collapse/mixture problems.Specifically, we first faithfully embed the low dimensionalimage manifold into the latent space by training an autoen-coder (AE). Then we compute the optimal transport (OT)map that pushes forward the uniform distribution to the la-tent distribution supported on the latent manifold. Finally,our GAN model is trained to generate high quality imagesfrom the latent distribution, the distribution transform mapfrom which to the empirical data distribution will be con-tinuous. The paired data between the latent code and thereal images gives us further constriction about the generator.Experiments on simple MNIST dataset and complex datasetslike Cifar-10 and CelebA show the efficacy and efficiency ofour proposed method.

cs.CV↗

arXiv2019

Demonstration of beta-Ga2O3 Optical Waveguides and the Analysis of Their Propagation Losses in the UV-Visible Spectra

This paper reports the first demonstration of beta-phase gallium oxide as optical waveguides on sapphire substrates grown by metal-organic chemical vapor deposition (MOCVD). The propagation losses from visible to ultraviolet spectra were comprehensively studied. By optimizing the fabrication processes, minimum propagation loss was identified to be 3.7 dB/cm at the wavelength of 810 nm, which is comparable to other wide bandgap materials within the III-N family (GaN, AlN). To further reveal the underlying loss mechanisms, several physical mechanisms such as two-photon absorption, sidewall scattering, top surface scattering, and bulk scattering were taken into consideration. The results obtained from this work suggest that beta-Ga2O3 is promising for ultraviolet-visible spectrum integrated photonic applications.

physics.optics↗

arXiv2019

Mode Collapse and Regularity of Optimal Transportation Maps

This work builds the connection between the regularity theory of optimal transportation map, Monge-Ampère equation and GANs, which gives a theoretic understanding of the major drawbacks of GANs: convergence difficulty and mode collapse. According to the regularity theory of Monge-Ampère equation, if the support of the target measure is disconnected or just non-convex, the optimal transportation mapping is discontinuous. General DNNs can only approximate continuous mappings. This intrinsic conflict leads to the convergence difficulty and mode collapse in GANs. We test our hypothesis that the supports of real data distribution are in general non-convex, therefore the discontinuity is unavoidable using an Autoencoder combined with discrete optimal transportation map (AE-OT framework) on the CelebA data set. The testing result is positive. Furthermore, we propose to approximate the continuous Brenier potential directly based on discrete Brenier theory to tackle mode collapse. Comparing with existing method, this method is more accurate and effective.

cs.LG↗

arXiv2018

Brain Morphometry Analysis with Surface Foliation Theory

Brain morphometry study plays a fundamental role in neuroimaging research. In this work, we propose a novel method for brain surface morphometry analysis based on surface foliation theory. Given brain cortical surfaces with automatically extracted landmark curves, we first construct finite foliations on surfaces. A set of admissible curves and a height parameter for each loop are provided by users. The admissible curves cut the surface into a set of pairs of pants. A pants decomposition graph is then constructed. Strebel differential is obtained by computing a unique harmonic map from surface to pants decomposition graph. The critical trajectories of Strebel differential decompose the surface into topological cylinders. After conformally mapping those topological cylinders to standard cylinders, parameters of standard cylinders (height, circumference) are intrinsic geometric features of the original cortical surfaces and thus can be used for morphometry analysis purpose. In this work, we propose a set of novel surface features rooted in surface foliation theory. To the best of our knowledge, this is the first work to make use of surface foliation theory for brain morphometry analysis. The features we computed are intrinsic and informative. The proposed method is rigorous, geometric, and automatic. Experimental results on classifying brain cortical surfaces between patients with Alzheimer's disease and healthy control subjects demonstrate the efficiency and efficacy of our method.

cs.CG↗

arXiv2018

Adversarial Domain Adaptation for Classification of Prostate Histopathology Whole-Slide Images

Automatic and accurate Gleason grading of histopathology tissue slides is crucial for prostate cancer diagnosis, treatment, and prognosis. Usually, histopathology tissue slides from different institutions show heterogeneous appearances because of different tissue preparation and staining procedures, thus the predictable model learned from one domain may not be applicable to a new domain directly. Here we propose to adopt unsupervised domain adaptation to transfer the discriminative knowledge obtained from the source domain to the target domain without requiring labeling of images at the target domain. The adaptation is achieved through adversarial training to find an invariant feature space along with the proposed Siamese architecture on the target domain to add a regularization that is appropriate for the whole-slide images. We validate the method on two prostate cancer datasets and obtain significant classification improvement of Gleason scores as compared with the baseline models.

cs.CV↗

arXiv2016

Effects of beam velocity and density on an ion-beam pulse moving in magnetized plasmas

The wakefield and stopping power of an ion-beam pulse moving in magnetized plasmas are investigated by particle-in-cell (PIC) simulations. The effects of beam velocity and density on the wake and stopping power are discussed. In the presence of magnetic field, it is found that beside the longitudinal conversed V-shaped wakes, the strong whistler wave are observed when low-density and low-velocity pulses moving in plasmas. The corresponding stopping powers are enhanced due to the drag of these whistler waves. As beam velocities increase, the whistler waves disappear, and only are conversed V-shape wakes observed. The corresponding stopping powers are reduced compared with these in isotropic plasmas. When high-density pulses transport in the magnetized plasmas, the whistler waves are greatly inhibited for low-velocity pulses and disappear for high-velocity pulses. Additionally, the magnetic field reduces the stopping powers for all high-density cases.

physics.plasm-ph↗

arXiv2015

Signal extraction approach for sparse multivariate response regression

In this paper, we consider multivariate response regression models with high dimensional predictor variables. One way to model the correlation among the response variables is through the low rank decomposition of the coefficient matrix, which has been considered by several papers for the high dimensional predictors. However, all these papers focus on the singular value decomposition of the coefficient matrix. Our target is the decomposition of the coefficient matrix which leads to the best lower rank approximation to the regression function, the signal part in the response. Given any rank, this decomposition has nearly the smallest expected prediction error among all approximations to the the coefficient matrix with the same rank. To estimate the decomposition, we formulate a penalized generalized eigenvalue problem to obtain the first matrix in the decomposition and then obtain the second one by a least squares method. In the high-dimensional setting, we establish the oracle inequalities for the estimates. Compared to the existing theoretical results, we have less restrictions on the distribution of the noise vector in each observation and allow correlations among its coordinates. Our theoretical results do not depend on the dimension of the multivariate response. Therefore, the dimension is arbitrary and can be larger than the sample size and the dimension of the predictor. Simulation studies and application to real data show that the proposed method has good prediction performance and is efficient in dimension reduction for various reduced rank models.

stat.ME↗

arXiv2015

Sparse Fisher's discriminant analysis with thresholded linear constraints

Various regularized linear discriminant analysis (LDA) methods have been proposed to address the problems of the classic methods in high-dimensional settings. Asymptotic optimality has been established for some of these methods in high dimension when there are only two classes. A major difficulty in proving asymptotic optimality for multiclass classification is that the classification boundary is typically complicated and no explicit formula for classification error generally exists when the number of classes is greater than two. For the Fisher's LDA, one additional difficulty is that the covariance matrix is also involved in the linear constraints. The main purpose of this paper is to establish asymptotic consistency and asymptotic optimality for our sparse Fisher's LDA with thresholded linear constraints in the high-dimensional settings for arbitrary number of classes. To address the first difficulty above, we provide asymptotic optimality and the corresponding convergence rates in high-dimensional settings for a large family of linear classification rules with arbitrary number of classes, and apply them to our method. To overcome the second difficulty, we propose a thresholding approach to avoid the estimate of the covariance matrix. We apply the method to the classification problems for multivariate functional data through the wavelet transformations.

stat.ME↗

arXiv2015

Asymptotic optimality of sparse linear discriminant analysis with arbitrary number of classes

Many sparse linear discriminant analysis (LDA) methods have been proposed to overcome the major problems of the classic LDA in high-dimensional settings. However, the asymptotic optimality results are limited to the case that there are only two classes, which is due to the fact that the classification boundary of LDA is a hyperplane and explicit formulas exist for the classification error in this case. In the situation where there are more than two classes, the classification boundary is usually complicated and no explicit formulas for the classification errors exist. In this paper, we consider the asymptotic optimality in the high-dimensional settings for a large family of linear classification rules with arbitrary number of classes under the situation of multivariate normal distribution. Our main theorem provides easy-to-check criteria for the asymptotic optimality of a general classification rule in this family as dimensionality and sample size both go to infinity and the number of classes is arbitrary. We establish the corresponding convergence rates. The general theory is applied to the classic LDA and the extensions of two recently proposed sparse LDA methods to obtain the asymptotic optimality. We conduct simulation studies on the extended methods in various settings.

math.ST↗

arXiv2010

Asymptotic efficiency and finite-sample properties of the generalized profiling estimation of parameters in ordinary differential equations

Ordinary differential equations (ODEs) are commonly used to model dynamic behavior of a system. Because many parameters are unknown and have to be estimated from the observed data, there is growing interest in statistics to develop efficient estimation procedures for these parameters. Among the proposed methods in the literature, the generalized profiling estimation method developed by Ramsay and colleagues is particularly promising for its computational efficiency and good performance. In this approach, the ODE solution is approximated with a linear combination of basis functions. The coefficients of the basis functions are estimated by a penalized smoothing procedure with an ODE-defined penalty. However, the statistical properties of this procedure are not known. In this paper, we first give an upper bound on the uniform norm of the difference between the true solutions and their approximations. Then we use this bound to prove the consistency and asymptotic normality of this estimation procedure. We show that the asymptotic covariance matrix is the same as that of the maximum likelihood estimation. Therefore, this procedure is asymptotically efficient. For a fixed sample and fixed basis functions, we study the limiting behavior of the approximation when the smoothing parameter tends to infinity. We propose an algorithm to choose the smoothing parameters and a method to compute the deviation of the spline approximation from solution without solving the ODEs.

math.ST↗