Ping Guo

Ping Guo
Beijing Normal University | bnu · School of Systems Science

PhD in Computer Science

About

379
Publications
67,925
Reads
How we measure 'reads'
A 'read' is counted each time someone views a publication summary (such as the title, abstract, and list of authors), clicks on a figure, or views or downloads the full-text. Learn more
2,978
Citations
Additional affiliations
January 2016 - August 2020
Beijing Normal University
Position
  • Professor (Full)
June 2001 - December 2015
Beijing Normal University
Position
  • Professor (Full)
December 1983 - May 1997
Beijing Normal University
Position
  • Teacher
Education
May 1997 - June 2001
The Chinese University of Hong Kong
Field of study
  • Computer Science & Engineering

Publications

Publications (379)
Preprint
Full-text available
While exogenous variables have a major impact on performance improvement in time series analysis, inter-series correlation and time dependence among them are rarely considered in the present continuous methods. The dynamical systems of multivariate time series could be modelled with complex unknown partial differential equations (PDEs) which play a...
Article
In this paper, a synergetic learning structure-based neuro-optimal fault tolerant control (SLSNOFTC) method is proposed for unknown nonlinear continuous-time systems with actuator failures. Under the framework of the synergetic learning structure (SLS), the optimal control input and the actuator failure are viewed as two subsystems. Then, the fault...
Article
Searching for rare astronomical objects based on spectral data is similar to finding needles in a haystack owing to their rarity and the immense data volume gathered from large astronomical spectroscopic surveys. In this paper, we propose a novel automated approximate nearest neighbor search method based on unsupervised hashing learning for rare sp...
Article
In this work, a recognition model with low computation and high training efficiency is proposed for images recognition. The proposed model consists of two modules: 1) haar wavelet transformation extracts features and compresses images; 2) a representation learning module using autoencoder. Because of its powerful function, Haar wavelet transformati...
Article
Full-text available
The success of deep learning in skin lesion classification mainly depends on the ultra-deep neural network and the significantly large training data set. Deep learning training is usually time-consuming, and large datasets with labels are hard to obtain, especially skin lesion images. Although pre-training and data augmentation can alleviate these...
Article
Full-text available
Deep neural networks offer advanced procedures for many learning tasks because of the ability to extract preferable features at every network layer. The evolved efficiency of extra layers inside a deep network will come at the expense of appended latency and power consumption in feedforward inference. As networks continue to grow and deepen, these...
Article
Full-text available
Autoencoder has been widely used as a feature learning technique. In many works of autoencoder, the features of the original input are usually extracted layer by layer using multi-layer nonlinear mapping, and only the features of the last layer are used for classification or regression. Therefore, the features of the previous layer aren’t used expl...
Article
Full-text available
In this paper, a sliding mode (SM)-based online fault compensation control scheme is investigated for modular reconfigurable robots (MRRs) with actuator failures via adaptive dynamic programming. It consists of a SM-based iterative controller, an adaptive robust term and an online fault compensator. For fault-free MRR systems, the SM surface-based...
Article
By training different models and averaging their predictions, the performance of the machine-learning algorithm can be improved. The performance optimization of multiple models is supposed to generalize further data well. This requires the knowledge transfer of generalization information between models. In this article, a multiple kernel mutual lea...
Article
Pseudo-inverse learners (PILs) are a kind of feedforward neural network trained with the pseudoinverse learning algorithm, which can be traced back to 1995 originally. PIL is an approach for nongradient descent learning, and its main advantage is the lower computational cost and fast learning procedure, which is especially relevant in the edge comp...
Article
Modeling the relevance matching between a query and a document is challenging in ad-hoc document retrieval due to diverse semantic matching aspects. Both explicit and implicit semantic matching aspects exist, corresponding to the query term occurrences in a document and the term transformations between queries and documents respectively. While most...
Article
The convolutional neural network is the most widely used deep neural network. However, it still has some disadvantages. First, the back-propagation method is usually used in the training of convolutional neural networks, but it has several inherent defects, such as the vanishing gradient problem and exploding gradient problem. Moreover, the trainin...
Article
Full-text available
Developing a robust deep neural network (DNN) for a specific task is not only time-consuming but also requires lots of experienced human experts. In order to make deep neural networks easier to apply or even take the human experts out of the design of network architecture completely, a growing number of researches focus on robust automated machine...
Preprint
Full-text available
In this work, we generalize the reaction-diffusion equation in statistical physics, Schr\"odinger equation in quantum mechanics, Helmholtz equation in paraxial optics into the neural partial differential equations (NPDE), which can be considered as the fundamental equations in the field of artificial intelligence research. We take finite difference...
Article
Full-text available
In the bioinformatics field, the classification of unknown biological sequences is a key task that is fundamental for simplifying the consistency, aggregation, and survey of organisms and their evolution. We can view biological sequences as data components of higher non-fixed dimensions, corresponding to the length of the sequences. Numerical encod...
Article
The studies on users’ purchase intentions based on e-commerce data are of great significance to marketers, buyers, and society. Current studies on users’ intentions with traditional machine learning methods usually focus on unique features and are time-consuming. Due to the characteristics of user behaviors and the importance of time sequence, deep...
Preprint
Full-text available
In this work, we generalize the reaction-diffusion equation in statistical physics, Schrödinger equation in quantum mechanics, Helmholtz equation in paraxial optics into the neural partial differential equations (PDE), which can be considered as the fundamental equations in the field of artificial intelligence research. We take finite difference me...
Chapter
Software aging, also called smooth degradation or chronics, has been observed in a long running software application, accompanied by performance degradation, hang/crash failures or both. The key for software aging problem is how to fast and accurately detect software aging occurrence, which is a hard work due to the long delay before aging appearan...
Preprint
Full-text available
In a two-models synergetic learning systems (SLS), one or both models is/are assigned to be the Pseudoinverse Learners (PILer). In this work, a more general framework of PIL's structure evolutionary is investigated, and extreme deep and extreme wide PILer architecture is proposed. Moreover, we present that the high order neural network such as rand...
Article
A pulsar is a rapidly rotating neutron star and transmits periodic oscillations of power to the earth. We introduce a novel method for pulsar candidate classification. The method contains two major steps: (1) make strong representations for pulsar candidate in the image domain by extracting deep features with the deep convolutional generative adver...
Article
The dual hesitant fuzzy set(DHFS) is a useful tool to deal with situations in which people are hesitant about providing their satisfaction degree and dissatisfaction degree. In this paper, we introduce the concepts of weighted dual hesitant fuzzy set(WDHFS) and weighted dual hesitant fuzzy element(WDHFE). Furthermore, we introduce some basic operat...
Chapter
Causal inference between two observed variables has received a widespread attention in science. Generally, most existing approaches are focusing on inferring the casual direction based on data of the same type. However, in practice, it is very common that the observations obtained from different measurements can have different data types. This issu...
Presentation
Full-text available
我们提出了将最小作用量原理作为人工智能的第一性原理;基于最小作用量原理,对系统的自由能变分,可导出描述AI系统的基本方程----反应-扩散方程。 薛定谔方程是量子物理的基本方程,麦克斯韦方程是电磁学的基本方程,而反应-扩散方程则是AI系统的基本方程之一。 反应-扩散方程(图灵方程)是耗散系统的基本方程。在数学上,反应-扩散方程是一类半线性抛物方程。在人工智能系统,反应-扩散方程是非线性抛物方程。
Article
Full-text available
Pest infestation of crops and plants impacts agricultural development. Generally, farmers or specialist observe the plants with the naked eye to recognise and diagnose ailments. However, this technique can be time-consuming, costly and inexact. In contrast, auto-detection using image processing methods gives fast and precise results. This paper int...
Article
Radio frequency interference (RFI) is an important challenge in radio astronomy. RFI comes from various sources and increasingly impacts astronomical observation as telescopes become more sensitive. In this study, we propose a fast and effective method for removing RFI in pulsar data. We use pseudo-inverse learning to train a single hidden layer au...
Method
Full-text available
In this paper, the first principle of artificial intelligence is explained by the expression of "seven questions". That is, 1. What is the first principles? 2. Why the first principle is used? 3. What is first principle thinking? 4. How to use the first principle? 5. Is there the first principles in the field of artificial intelligence? 6. Why the...
Preprint
Full-text available
Drawing on the idea that brain development is a Darwinian process of ``evolution + selection'' and the idea that the current state is a local equilibrium state of many bodies with self-organization and evolution processes driven by the temperature and gravity in our universe, in this work, we describe an artificial intelligence system called the ``...
Article
This paper presents an ensemble classification model with unsupervised feature learning for driving stress recognition under real-world driving conditions. The driving stress is detected using drivers' different physiological signals, specifically the electromyogram, electrocardiogram, galvanic skin response, heart rate and respiration. The propose...
Preprint
Due to limited size and imperfect of the optical components in a spectrometer, aberration has inevitably been brought into two-dimensional multi-fiber spectrum image in LAMOST, which leads to obvious spacial variation of the point spread functions (PSFs). Consequently, if spatial variant PSFs are estimated directly , the huge storage and intensive...
Article
Full-text available
The main objective of blind image de-blurring is to recover a sharp image from a given blurry image. A good estimation of the kernel plays an important role in recovering a sharp image. However, if the local object textures are neglected when the kernel is being estimated, this can lead to over-smoothing or can produce a strong ringing effect. In t...
Article
Discovering pulsars is a significant and meaningful research topic in the field of radio astronomy. With the advent of astronomical instruments, the volume and rate of data acquisition have grown exponentially. This development necessitates a focus on artificial intelligence (AI) technologies that can mine large astronomical data sets. Automatic pu...
Article
Software systems have become larger and more complex than ever. Such characteristics make it very challengeable to prevent software defects. Therefore, automatically predicting the number of defects in software modules is necessary and may help developers efficiently to allocate limited resources. Various approaches have been proposed to identify a...
Article
Full-text available
The multi-target fiber spectroscopic telescope can obtain a large number of spectral data of different celestial bodies in one observation. The light detected from the celestial body passes through the slit of the spectrometer, and after passing through the optical fiber, it is transmitted to the CCD sensor to obtain a two-dimensional spectral imag...
Article
Full-text available
In addition to the spectral redshift of galaxies, the prediction of galaxies redshift has important research significance for studying the large-scale structure and evolution of the universe. In this paper, we use the metering and spectral data of 150 000 galaxies of SDSS DR13 released by the Sloan Sky Survey project to analyze the galaxies accordi...
Article
Due to its flexibility and good restoration performance, the Expected Patch Log Likelihood (EPLL) method has attracted extensive attention and has been further developed. However, the basic EPLL method is mainly applied for gray image restoration. For color image denoising with different channel noise levels, concatenating the RGB values into a vec...
Chapter
In this paper, an unified view on feedforward neural networks (FNNs) is provided from the free perception of the architecture design, learning algorithm, cost function, regularization, activation functions, etc. Furthermore, we consider the ensemble networks and swarm intelligence as the same sytems which consists of multiple learning algorithms or...
Article
Full-text available
This paper centers on a novel method for traffic sign recognition (TSR). The method comprises of two major steps: 1) make strong representations for TSR images, by extraction deep features with the Deep Convolutional Generative Adversarial Networks (DCGAN); 2) classifier defined by Multilayer Perceptron (MLP) neural networks trained with Pseudoinve...
Chapter
In this paper, a gradient-descent neurodynamic approach is proposed for the distributed linear programming problem with affine equality constraints. It is rigorously proved that the state solution of the proposed gradient-descent approach with an arbitrary initial point reaches agreement and is convergent to an optimal solution of the considered op...
Data
This is the PPT presented at China Systems Science Conference (CSSC 2019). May 18-19, 2019. Chang Sha, Hunan Province, China.
Preprint
Full-text available
This paper reviews the development history of artificial intelligence, analyzed the current research status at our country and worldwide. And points out the dilemma of basic research of artificial intelligence represented by deep learning. It includes interpretable neural network model problem, network architecture design problem, small sample lear...
Preprint
Full-text available
In this paper, based on the free energy principle, the statistical physics interpretation of the neural network is given, and the interpretability problem of deep neural model is investigated. Our proposed synergetic learning system is an artificial intelligence system with multiple subsystems. One of the special cases is a model containing two sub...
Article
The Commensal Radio Astronomy Five-hundred-meter Aperture Spherical radio Telescope (FAST) Survey (CRAFTS) utilizes the novel drift-scan commensal survey mode of FAST and can generate billions of pulsar candidate signals. The human experts are not likely to thoroughly examine these signals, and various machine sorting methods are used to aid the cl...
Chapter
Deep convolution neural network (CNN) is one of the most popular Deep neural networks (DNN). It has won state-of-the-art performance in many computer vision tasks. The most used method to train DNN is Gradient descent-based algorithm such as Backpropagation. However, backpropagation algorithm usually has the problem of gradient vanishing or gradien...
Chapter
Full-text available
Pseudoinverse learner (PIL), a kind of multilayer neural networks (MLP) trained with pseudoinverse learning algorithm, is a novel learning framework. It has drawn increasing attention in the areas of large-scale computing, high-speed signal processing, artificial intelligence and so on. In this paper, we briefly review the pseudoinverse learning al...
Preprint
Full-text available
The Commensal Radio Astronomy Five-hundred-meter Aperture Spherical radio Telescope (FAST) Survey (CRAFTS) utilizes the novel drift-scan commensal survey mode of FAST and can generate billions of pulsar candidate signals. The human experts are not likely to thoroughly examine these signals, and various machine sorting methods are used to aid the cl...
Preprint
Full-text available
Synergetic learning system is an artificial intelligence system. In this paper the methodology of building a synergetic learning system is presented. It is proposed that the synergetic learning system should be analyzed and studied systematically from the view point of the multi-scale, multi-level, and multi-perspective. Drawing on the mathematical...
Preprint
Full-text available
Drawing on the idea that brain development is a Darwinian process of ``evolution + selection'', and the idea that current state is a local equilibrium state of many bodies with self-organization and evolution process driven by the temperature and gravity in our Universe, in this work, we describe an artificial intelligence system called ``Synergeti...
Conference Paper
On-line news reading has become the most popular way for user to obtain real-time information. With the millions of news, it is a key challenge to help user find the articles that are interesting to read. Although great achievements have been made, there is little work to focus on combing news language with external knowledge graphs and expanding n...
Preprint
Full-text available
In this work, a non-gradient descent learning scheme is proposed for deep feedforward neural networks (DNN). As we known, autoencoder can be used as the building blocks of the multi-layer perceptron (MLP) deep neural network. So, the MLP will be taken as an example to illustrate the proposed scheme of pseudoinverse learning algorithm for autoencode...
Article
Fingerprint recognition systems are widely used for authentication purposes in security systems. However, fingerprint recognition systems can easily be spoofed by imitations of fingerprints using various spoof materials. A compact and discriminative set of features is needed to discriminate between live and spoof fingerprints. We explore combined S...
Conference Paper
Autoencoder is one approach to automatically learn features from unlabeled data and received significant attention during the development of deep neural networks. However, the learning algorithm of autoencoder suffers from slow learning speed because of gradient descent based algorithms have many drawbacks. Pseudoinverse learning algorithm is a fas...
Preprint
Full-text available
The objective of this experiment report is to investigate the method of pulsar candidate classification with a generative adversarial network. Two goals included: 1) to validate that a generative adversarial network is able to produce a lot of new fake but realistic positive examples which would be helpful for solving the class imbalance problem in...
Presentation
Full-text available
In this presentation, I give the unified feedforward neural network architecture design framework, which includes traditional MLP, ResNet, Deep Convex Network, as well as other variants. The basic building block is PseudoInverse Learner (PILer), and PILer has various different architectures.
Chapter
In this paper, one novel method to extract flux from two dimensional spectral images which we observed through LAMOST (Large Area Multi-Object Fiber Spectroscopic Telescope) is proposed. First of all, the spectral images are preprocessed. Then, in the flux extraction algorithm, the GRNN (General Regression Neural Network) and double Gaussian functi...
Chapter
Natural language generation is a critical component of dialogue system and plenty of works have proved the effectiveness and efficiency of sequence-to-sequence (seq2seq) model for generation. Seq2seq model is a kind of neural networks which usual require massive data to learn its parameters. For many small shops in customer service dialogue systems...
Chapter
In this work, we give an overview of pseudoinverse learning (PIL) algorithm as well as applications. PIL algorithm is a non-gradient descent algorithm for multi-layer perception. The weight matrix of network can be exactly computed by PIL algorithm. So PIL algorithm can effectively avoid the problem of low convergence and local minima. Moreover, PI...