arXiv subjects

Yi Cheng

Publications and source records attributed to Yi Cheng.

At least 91 records · Page 5Linked to original sources

arXiv2022

Research on Parallel SVM Algorithm Based on Cascade SVM

Cascade SVM (CSVM) can group datasets and train subsets in parallel, which greatly reduces the training time and memory consumption. However, the model accuracy obtained by using this method has some errors compared with direct training. In order to reduce the error, we analyze the causes of error in grouping training, and summarize the grouping without error under ideal conditions. A Balanced Cascade SVM (BCSVM) algorithm is proposed, which balances the sample proportion in the subset after grouping to ensure that the sample proportion in the subset is the same as the original dataset. At the same time, it proves that the accuracy of the model obtained by BCSVM algorithm is higher than that of CSVM. Finally, two common datasets are used for experimental verification, and the results show that the accuracy error obtained by using BCSVM algorithm is reduced from 1% of CSVM to 0.1%, which is reduced by an order of magnitude.

cs.LG↗

arXiv2021

Unifying Discourse Resources with Dependency Framework

For text-level discourse analysis, there are various discourse schemes but relatively few labeled data, because discourse research is still immature and it is labor-intensive to annotate the inner logic of a text. In this paper, we attempt to unify multiple Chinese discourse corpora under different annotation schemes with discourse dependency framework by designing semi-automatic methods to convert them into dependency structures. We also implement several benchmark dependency parsers and research on how they can leverage the unified data to improve performance.

cs.CL↗

arXiv2021

Guiding the Growth: Difficulty-Controllable Question Generation through Step-by-Step Rewriting

This paper explores the task of Difficulty-Controllable Question Generation (DCQG), which aims at generating questions with required difficulty levels. Previous research on this task mainly defines the difficulty of a question as whether it can be correctly answered by a Question Answering (QA) system, lacking interpretability and controllability. In our work, we redefine question difficulty as the number of inference steps required to answer it and argue that Question Generation (QG) systems should have stronger control over the logic of generated questions. To this end, we propose a novel framework that progressively increases question difficulty through step-by-step rewriting under the guidance of an extracted reasoning chain. A dataset is automatically constructed to facilitate the research, on which extensive experiments are conducted to test the performance of our method.

cs.CL↗

arXiv2021

6D Pose Estimation with Correlation Fusion

6D object pose estimation is widely applied in robotic tasks such as grasping and manipulation. Prior methods using RGB-only images are vulnerable to heavy occlusion and poor illumination, so it is important to complement them with depth information. However, existing methods using RGB-D data cannot adequately exploit consistent and complementary information between RGB and depth modalities. In this paper, we present a novel method to effectively consider the correlation within and across both modalities with attention mechanism to learn discriminative and compact multi-modal features. Then, effective fusion strategies for intra- and inter-correlation modules are explored to ensure efficient information flow between RGB and depth. To our best knowledge, this is the first work to explore effective intra- and inter-modality fusion in 6D pose estimation. The experimental results show that our method can achieve the state-of-the-art performance on LineMOD and YCB-Video dataset. We also demonstrate that the proposed method can benefit a real-world robot grasping task by providing accurate object pose estimation.

cs.CV↗

arXiv2021

Sound Trapping in an Open Resonator

The ability of extreme sound energy confinement with high-quality factor (Q-factor) resonance is of vital importance for acoustic devices requiring high intensity and hypersensitivity in biological ultrasonics, enhanced collimated sound emission (i.e. sound laser) and high-resolution sensing. However, structures reported so far demonstrated a limited quality factor (Q-factor) of acoustic resonances, up to several tens in an open resonator. The emergence of bound states in the continuum (BIC) makes it possible to realize high-Q factor acoustic modes. Here, we report the theoretical design and experimental demonstration of acoustic BICs supported by a single open resonator. We predicted that such an open acoustic resonator could simultaneously support three types of BICs, including symmetry protected BIC, Friedrich-Wintgen BIC induced by mode interference, as well as a new kind of BIC: mirror-symmetry induced BIC. We also experimentally demonstrated the existence of all three types of BIC with Q-factor up to one order of magnitude greater than the highest Q-factor reported in an open resonator.

physics.app-ph↗

arXiv2020

Rogue waves and lumps on the non-zero background in the PT -symmetric nonlocal Maccari system

In this paper, the PT -symmetric version of the Maccari system is introduced, which can be regarded as a two-dimensional generalization of the defocusing nonlocal nonlinear Schrodinger equation. Various exact solutions of the nonlocal Maccari system are obtained by means of the Hirota bilinear method, long-wave limit, and Kadomtsev-Petviashvili (KP) hierarchy method. Bilinear forms of the nonlocal Maccari system are derived for the first time. Simultaneously, a new nonlocal Davey-Stewartson-type equation is derived. Solutions for breathers and breathers on top of periodic line waves are obtained through the bilinear form of the nonlocal Maccari system. Hyperbolic line rogue-wave solutions and semi-rational ones, composed of hyperbolic line rogue wave and periodic line waves are also derived in the long-wave limit. The semi-rational solutions exhibit a unique dynamical behavior. Additionally, general line soliton solutions on constant background are generated by restricting different tau-functions of the KP hierarchy, combined with the Hirota bilinear method. These solutions exhibit elastic collisions, some of which have never been reported before in nonlocal systems. Additionally, the semi-rational solutions, namely (i) fusion of line solitons and lumps into line solitons, and (ii) fission of line solitons into lumps and line solitons, are put forward in terms of the KP hierarchy. These novel semi-rational solutions reduce to 2N-lump solutions of the nonlocal Maccari system with appropriate parameters. Finally, different characteristics of exact solutions for the nonlocal Maccari system are summarized. These new results enrich the structure of waves in nonlocal nonlinear systems, and help to understand new physical phenomena.

nlin.SI↗

arXiv2020

Deformed two-dimensional rogue waves in the (2+1)-dimensional Korteweg-de Vries equation

Within the (2 + 1)-dimensional Korteweg-de Vries equation framework, new bilinear Backlund transformation and Lax pair are presented based on the binary Bell polynomials and gauge transformation. By introducing an arbitrary function, a family of deformed soliton and deformed breather solutions are presented with the improved Hirotas bilinear method. Choosing the appropriate parameters, their interesting dynamic behaviors are shown in three-dimensional plots. Furthermore, novel rational solutions are generated by taking the limit of obtained solitons. Additionally, two dimensional [2D] rogue waves (localized in both space and time) on the soliton plane are presented, we refer to it as deformed 2D rogue waves. The obtained deformed 2D rogue waves can be viewed as a 2D analog of the Peregrine soliton on soliton plane, and its evolution process is analyzed in detail. The deformed 2D rogue wave solutions are constructed successfully, which are closely related to the arbitrary function. This new idea is also applicable to other nonlinear systems.

nlin.SI↗

arXiv2020

Duality and sensitivity analysis of multistage linear stochastic programs

We investigate the dual of a Multistage Stochastic Linear Program (MSLP) to study two questions for this class of problems. The first of these questions is the study of the optimal value of the problem as a function of the involved parameters. For this sensitivity analysis problem, we provide formulas for the derivatives of the value function with respect to the parameters and illustrate their application on an inventory problem. Since these formulas involve optimal dual solutions, we need an algorithm that computes such solutions to use them. In this context, the second question we address is the study of solution methods for the dual problem. Writing Dynamic Programming (DP) equations for the dual, we can use an SDDP type method, called Dual SDDP, which solves these DP equations. However, applying this method will only be possible if the Relatively Complete Recourse (RCR) holds for the dual. Since the RCR assumption may fail to hold, we design two variants of Dual SDDP, namely Dual SDDP with penalizations and Dual SDDP with feasibility cuts. The study of this second question allows us to take a fresh look at the notoriously difficult to solve class of MSLP with interstage dependent cost coefficients. Indeed, for this class of problems, cost-to-go functions are non-convex and solution methods were so far using SDDP for a Markov chain approximation of the cost coefficients process. For these problems, we propose to apply Dual SDDP with penalizations to the cost-to-go functions of the dual which are concave. This algorithm converges to the optimal value of the problem. Finally, as a proof of concept of the tools developed, we present the results of numerical experiments computing the sensitivity of the optimal value of an inventory problem as a function of parameters of the demand process and compare Primal and Dual SDDP on the inventory and a hydro-thermal planning problems.

math.OC↗

arXiv2020

Reductions of the (4 + 1)-dimensional Fokas equation and their solutions

An integrable extension of the Kadomtsev-Petviashvili (KP) and Davey-Stewartson (DS) equations is investigated in this paper.We will refer to this integrable extension as the (4+1)-dimensional Fokas equation. The determinant expressions of soliton, breather, rational, and semi-rational solutions of the (4 + 1)-dimensional Fokas equation are constructed based on the Hirota's bilinear method and the KP hierarchy reduction method. The complex dynamics of these new exact solutions are shown in both three-dimensional plots and two-dimensional contour plots. Interestingly, the patterns of obtained high-order lumps are similar to those of rogue waves in the (1 + 1)-dimensions by choosing different values of the free parameters of the model. Furthermore, three kinds of new semi-rational solutions are presented and the classification of lump fission and fusion processes is also discussed. Additionally, we give a new way to obtain rational and semi-rational solutions of (3 + 1)-dimensional KP equation by reducing the solutions of the (4 + 1)-dimensional Fokas equation. All these results show that the (4 + 1)-dimensional Fokas equation is a meaningful multidimensional extension of the KP and DS equations. The obtained results might be useful in diverse fields such as hydrodynamics, non-linear optics and photonics, ion-acoustic waves in plasmas, matter waves in Bose-Einstein condensates, and sound waves in ferromagnetic media.

nlin.SI↗

arXiv2019

Zero-shot Chinese Discourse Dependency Parsing via Cross-lingual Mapping

Due to the absence of labeled data, discourse parsing still remains challenging in some languages. In this paper, we present a simple and efficient method to conduct zero-shot Chinese text-level dependency parsing by leveraging English discourse labeled data and parsing techniques. We first construct the Chinese-English mapping from the level of sentence and elementary discourse unit (EDU), and then exploit the parsing results of the corresponding English translations to obtain the discourse trees for the Chinese text. This method can automatically conduct Chinese discourse parsing, with no need of a large scale of Chinese labeled data.

cs.CL↗

arXiv2019

Two-dimensional rogue waves on zero background of the Davey-Stewartson II equation

A prototypical example of a rogue wave structure in a two-dimensional model is presented in the context of the Davey-Stewartson~II (DS~II) equation arising in water waves. The analytical methodology involves a Taylor expansion of an eigenfunctionof the model's Lax pair which is used to form a hierarchy of infinitely many new eigenfunctions. These are used for the construction of two-dimensional (2D) rogue waves (RWs) of the DS~II equation by the even-fold Darboux transformation (DT). The obtained 2D RWs, which are localized in both space and time, can be viewed as a 2D analogue of the Peregrine soliton and are thus natural candidates to describe oceanic RW phenomena,as well as ones in 2D fluid systems and water tanks.

nlin.SI↗

arXiv2018

Look Across Elapse: Disentangled Representation Learning and Photorealistic Cross-Age Face Synthesis for Age-Invariant Face Recognition

Despite the remarkable progress in face recognition related technologies, reliably recognizing faces across ages still remains a big challenge. The appearance of a human face changes substantially over time, resulting in significant intra-class variations. As opposed to current techniques for age-invariant face recognition, which either directly extract age-invariant features for recognition, or first synthesize a face that matches target age before feature extraction, we argue that it is more desirable to perform both tasks jointly so that they can leverage each other. To this end, we propose a deep Age-Invariant Model (AIM) for face recognition in the wild with three distinct novelties. First, AIM presents a novel unified deep architecture jointly performing cross-age face synthesis and recognition in a mutual boosting way. Second, AIM achieves continuous face rejuvenation/aging with remarkable photorealistic and identity-preserving properties, avoiding the requirement of paired data and the true age of testing samples. Third, we develop effective and novel training strategies for end-to-end learning the whole deep architecture, which generates powerful age-invariant face representations explicitly disentangled from the age variation. Moreover, we propose a new large-scale Cross-Age Face Recognition (CAFR) benchmark dataset to facilitate existing efforts and push the frontiers of age-invariant face recognition research. Extensive experiments on both our CAFR and several other cross-age datasets (MORPH, CACD and FG-NET) demonstrate the superiority of the proposed AIM model over the state-of-the-arts. Benchmarking our model on one of the most popular unconstrained face recognition datasets IJB-C additionally verifies the promising generalizability of AIM in recognizing faces in the wild.

cs.CV↗

arXiv2018

Optimal Adaptive and Accelerated Stochastic Gradient Descent

Stochastic gradient descent (\textsc{Sgd}) methods are the most powerful optimization tools in training machine learning and deep learning models. Moreover, acceleration (a.k.a. momentum) methods and diagonal scaling (a.k.a. adaptive gradient) methods are the two main techniques to improve the slow convergence of \textsc{Sgd}. While empirical studies have demonstrated potential advantages of combining these two techniques, it remains unknown whether these methods can achieve the optimal rate of convergence for stochastic optimization. In this paper, we present a new class of adaptive and accelerated stochastic gradient descent methods and show that they exhibit the optimal sampling and iteration complexity for stochastic optimization. More specifically, we show that diagonal scaling, initially designed to improve vanilla stochastic gradient, can be incorporated into accelerated stochastic gradient descent to achieve the optimal rate of convergence for smooth stochastic optimization. We also show that momentum, apart from being known to speed up the convergence rate of deterministic optimization, also provides us new ways of designing non-uniform and aggressive moving average schemes in stochastic optimization. Finally, we present some heuristics that help to implement adaptive accelerated stochastic gradient descent methods and to further improve their practical performance for machine learning and deep learning.

stat.ML↗

arXiv2018

A Good Practice Towards Top Performance of Face Recognition: Transferred Deep Feature Fusion

Unconstrained face recognition performance evaluations have traditionally focused on Labeled Faces in the Wild (LFW) dataset for imagery and the YouTubeFaces (YTF) dataset for videos in the last couple of years. Spectacular progress in this field has resulted in saturation on verification and identification accuracies for those benchmark datasets. In this paper, we propose a unified learning framework named Transferred Deep Feature Fusion (TDFF) targeting at the new IARPA Janus Benchmark A (IJB-A) face recognition dataset released by NIST face challenge. The IJB-A dataset includes real-world unconstrained faces from 500 subjects with full pose and illumination variations which are much harder than the LFW and YTF datasets. Inspired by transfer learning, we train two advanced deep convolutional neural networks (DCNN) with two different large datasets in source domain, respectively. By exploring the complementarity of two distinct DCNNs, deep feature fusion is utilized after feature extraction in target domain. Then, template specific linear SVMs is adopted to enhance the discrimination of framework. Finally, multiple matching scores corresponding different templates are merged as the final results. This simple unified framework exhibits excellent performance on IJB-A dataset. Based on the proposed approach, we have submitted our IJB-A results to National Institute of Standards and Technology (NIST) for official evaluation. Moreover, by introducing new data and advanced neural architecture, our method outperforms the state-of-the-art by a wide margin on IJB-A dataset.

cs.CV↗

arXiv2017

Rational and semi-rational solutions of the nonlocal Davey-Stewartson equations

In this paper, the partially party-time ($PT$) symmetric nonlocal Davey-Stewartson (DS) equations with respect to $x$ is called $x$-nonlocal DS equations, while a fully $PT$ symmetric nonlocal DSII equation is called nonlocal DSII equation. Three kinds of solutions, namely breather, rational and semi-rational solutions for these nonlocal DS equations are derived by employing the bilinear method. For the $x$-nonlocal DS equations, the usual ($2+1$)-dimensional breathers are periodic in $x$ direction and localized in $y$ direction. Nonsingular rational solutions are lumps, and semi-rational solutions are composed of lumps, breathers and periodic line waves. For the nonlocal DSII equation, line breathers are periodic in both $x$ and $y$ directions with parallels in profile, but localized in time. Nonsingular rational solutions are ($2+1$)-dimensional line rogue waves, which arise from a constant background and disappear into the same constant background, and this process only lasts for a short period of time. Semi-rational solutions describe interactions of line rogue waves and periodic line waves.

nlin.SI↗

arXiv2016

Rogue wave triggered at a critical frequency of a nonlinear resonant medium

We consider a two-level atomic system, interacting with an electromagnetic field controlled in amplitude and frequency by a high intensity laser. We show that the amplitude of the induced electric field, admits an envelope profile corresponding to a breather soliton. We demonstrate that this soliton can propagate with any frequency shift with respect to that of the control laser, except a critical frequency, at which the system undergoes a structural discontinuity that transforms the breather in a rogue wave. A mechanism of generation of rogue waves by means of an intense laser field is thus revealed.

nlin.PS↗

arXiv2016

Event-triggered leader-following tracking control for multivariable multi-agent systems

The paper considers event-triggered leader-follower tracking control for multi-agent systems with general linear dynamics. For both undirected and directed follower graphs, we propose event triggering rules which guarantee bounded tracking errors. With these rules, we also prove that the systems do not exhibit Zeno behavior, and the bounds on the tracking errors can be tuned to a desired small value. We also show that the combinational state required for the proposed event triggering conditions can be continuously generated from discrete communications between the neighboring agents occurring at event times. The efficacy of the proposed methods is discussed using a simulation example.

eess.SY↗

arXiv2015

Leader-follower Tracking Control with Guaranteed Consensus Performance for Interconnected Systems with Linear Dynamic Uncertain Coupling

This paper considers the leader-follower tracking control problem for linear interconnected systems with undirected topology and linear dynamic coupling. Interactions between the systems are treated as linear dynamic uncertainty and are described in terms of integral quadratic constraints (IQCs). A consensus-type tracking control protocol is proposed for each system based on its state relative its neighbors. In addition a selected set of subsystems uses for control their relative states with respect to the leader. Two methods are proposed for the design of this control protocol. One method uses a coordinate transformation to recast the protocol design problem as a decentralized robust control problem for an auxiliary interconnected large scale system. Another method is direct, it does not employ coordinate transformation; it also allows for more general linear uncertain interactions. Using these methods, sufficient conditions are obtained which guarantee that the system tracks the leader. These conditions guarantee a suboptimal bound on the system consensus and tracking performance. The proposed methods are compared using a simulation example, and their effectiveness is discussed. Also, algorithms are proposed for computing suboptimal controllers.

eess.SY↗