Activity recognition plays a key role in health management and security.Traditional approaches are based on vision or wearables,which only work under the line of sight(LOS)or require the targets to carry dedicated dev...Activity recognition plays a key role in health management and security.Traditional approaches are based on vision or wearables,which only work under the line of sight(LOS)or require the targets to carry dedicated devices.As human bodies and their movements have influences on WiFi propagation,this paper proposes the recognition of human activities by analyzing the channel state information(CSI)from the WiFi physical layer.The method requires only the commodity:WiFi transmitters and receivers that can operate through a wall,under LOS and non-line of sight(NLOS),while the targets are not required to carry dedicated devices.After collecting CSI,the discrete wavelet transform is applied to reduce the noise,followed by outlier detection based on the local outlier factor to extract the activity segment.Activity recognition is fulfilled by using the bi-directional long short-term memory that takes the sequential features into consideration.Experiments in through-the-wall environments achieve recognition accuracy>95%for six common activities,such as standing up,squatting down,walking,running,jumping,and falling,outperforming existing work in this field.展开更多
The knowledge graph(KG) that represents structural relations among entities has become an increasingly important research field for knowledge-driven artificial intelligence. In this survey, a comprehensive review of K...The knowledge graph(KG) that represents structural relations among entities has become an increasingly important research field for knowledge-driven artificial intelligence. In this survey, a comprehensive review of KG and KG reasoning is provided. It introduces an overview of KGs, including representation, storage, and essential technologies. Specifically, it summarizes several types of knowledge reasoning approaches, including logic rules-based, representation-based, and neural network-based methods. Moreover, this paper analyzes the representation methods of knowledge hypergraphs. To effectively model hyper-relational data and improve the performance of knowledge reasoning, a three-layer knowledge hypergraph model is proposed. Finally, it analyzes the advantages of three-layer knowledge hypergraphs through reasoning and update algorithms which could facilitate future research.展开更多
In this paper, we approach the design of ID caching technology(IDCT) for graph databases, with the purpose of accelerating the queries on graph database data and avoiding redundant graph database query operations whic...In this paper, we approach the design of ID caching technology(IDCT) for graph databases, with the purpose of accelerating the queries on graph database data and avoiding redundant graph database query operations which will consume great computer resources. Traditional graph database caching technology(GDCT)needs a large memory to store data and has the problems of serious data consistency and low cache utilization. To address these issues, in the paper we propose a new technology which focuses on ID allocation mechanism and high-speed queries of ID on graph databases. Specifically, ID of the query result is cached in memory and data consistency is achieved through the real-time synchronization and cache memory adaptation. In addition, we set up complex queries and simple queries to satisfy all query requirements and design a mechanism of cache replacement based on query action time, query times, and memory capacity, thus improving the performance furthermore.Extensive experiments show the superiority of our techniques compared with the traditional query approach of graph databases.展开更多
Indoor floor plans are of vital importance for a wide range of indoor social applications, however, they are often unavailable due to various reasons. This paper proposes a method to automatically construct the indoor...Indoor floor plans are of vital importance for a wide range of indoor social applications, however, they are often unavailable due to various reasons. This paper proposes a method to automatically construct the indoor floor plan with rooms and corridors as well as landmarks, using inertial traces collected with smartphones. Landmarks,such as turnings, doors, and stairs, are identified according to inertial sensors and WiFi signals. The inertial traces are then partitioned into segments according to the turnings and classified as room type or corridor type according to the doors. Clustering is applied on room type and corridor type trace segments separately to produce room type and corridor type clusters. The construction of room applies the α-shape algorithm on room type clusters and the construction of corridor employs the principal component analysis(PCA) algorithm or the α-shape algorithm on corridor type clusters. Evaluations in two representative scenarios show that the method can construct the floor plan of acceptable accuracy with a relatively small set of inertial traces.展开更多
Several recent successes in deep learning(DL),such as state-of-the-art performance on several image classification benchmarks,have been achieved through the improved configuration.Hyperparameters(HPs)tuning is a key f...Several recent successes in deep learning(DL),such as state-of-the-art performance on several image classification benchmarks,have been achieved through the improved configuration.Hyperparameters(HPs)tuning is a key factor affecting the performance of machine learning(ML)algorithms.Various state-of-the-art DL models use different HPs in different ways for classification tasks on different datasets.This manuscript provides a brief overview of learning parameters and configuration techniques to show the benefits of using a large-scale handdrawn sketch dataset for classification problems.We analyzed the impact of different learning parameters and toplayer configurations with batch normalization(BN)and dropouts on the performance of the pre-trained visual geometry group 19(VGG-19).The analyzed learning parameters include different learning rates and momentum values of two different optimizers,such as stochastic gradient descent(SGD)and Adam.Our analysis demonstrates that using the SGD optimizer and learning parameters,such as small learning rates with high values of momentum,along with both BN and dropouts in top layers,has a good impact on the sketch image classification accuracy.展开更多
Hyperparameters are important for machine learning algorithms since they directly control the behaviors of training algorithms and have a significant effect on the performance of machine learning models. Several techn...Hyperparameters are important for machine learning algorithms since they directly control the behaviors of training algorithms and have a significant effect on the performance of machine learning models. Several techniques have been developed and successfully applied for certain application domains. However, this work demands professional knowledge and expert experience. And sometimes it has to resort to the brute-force search.Therefore, if an efficient hyperparameter optimization algorithm can be developed to optimize any given machine learning method, it will greatly improve the efficiency of machine learning. In this paper, we consider building the relationship between the performance of the machine learning models and their hyperparameters by Gaussian processes. In this way, the hyperparameter tuning problem can be abstracted as an optimization problem and Bayesian optimization is used to solve the problem. Bayesian optimization is based on the Bayesian theorem. It sets a prior over the optimization function and gathers the information from the previous sample to update the posterior of the optimization function. A utility function selects the next sample point to maximize the optimization function.Several experiments were conducted on standard test datasets. Experiment results show that the proposed method can find the best hyperparameters for the widely used machine learning models, such as the random forest algorithm and the neural networks, even multi-grained cascade forest under the consideration of time cost.展开更多
In the era of Big Data, typical architecture of distributed real-time stream processing systems is the combination of Flume, Kafka, and Storm. As a kind of distributed message system, Kafka has the characteristics of ...In the era of Big Data, typical architecture of distributed real-time stream processing systems is the combination of Flume, Kafka, and Storm. As a kind of distributed message system, Kafka has the characteristics of horizontal scalability and high throughput, which is manly deployed in many areas in order to address the problem of speed mismatch between message producers and consumers. When using Kafka, we need to quickly receive data sent by producers. In addition, we need to send data to consumers quickly. Therefore, the performance of Kafka is of critical importance to the performance of the whole stream processing system. In this paper, we propose the improved design of real-time stream processing systems, and focus on improving the Kafka's data loading process.We use Kafka cat to transfer data from the source to Kafka topic directly, which can reduce the network transmission. We also utilize the memory file system to accelerate the process of data loading, which can address the bottleneck and performance problems caused by disk I/O. Extensive experiments are conducted to evaluate the performance, which show the superiority of our improved design.展开更多
Three-dimensional(3D)shape registration is a challenging problem,especially for shapes under non-rigid transformations.In this paper,a 3D non-rigid shape registration method is proposed,called balanced functional maps...Three-dimensional(3D)shape registration is a challenging problem,especially for shapes under non-rigid transformations.In this paper,a 3D non-rigid shape registration method is proposed,called balanced functional maps(BFM).The BFM algorithm generalizes the point-based correspondence to functions.By choosing the Laplace-Beltrami eigenfunctions as the function basis,the transformations between shapes can be represented by the functional map(FM)matrix.In addition,many constraints on shape registration,such as the feature descriptor,keypoint,and salient region correspondence,can be formulated linearly using the matrix.By bi-directionally searching for the nearest neighbors of points’indicator functions in the function space,the point-based correspondence can be derived from FMs.We conducted several experiments on the Topology and Orchestration Specification for Cloud Applications(TOSCA)dataset and the Shape Completion and Animation of People(SCAPE)dataset.Experimental results show that the proposed BFM algorithm is effective and has superior performance than the state-of-the-art methods on both datasets.展开更多
This paper explores a way of deploying the classical algorithm named genetic algorithm(GA) with the memristor. The memristor is a type of circuit device with both characteristics of storage and computing, which provid...This paper explores a way of deploying the classical algorithm named genetic algorithm(GA) with the memristor. The memristor is a type of circuit device with both characteristics of storage and computing, which provides the similarity between electronic devices and biological components, such as neurons, and the structure of the memristor-based array is similar to that of chromosomes in genetics. Besides, it provides the similarity to the image gray-value matrix that can be applied to image restoration with GA. Thus, memristor-based GA is proposed and the experiment about image restoration using memristor-based GA is carried out in this paper. And parameters,such as the size of initial population and the number of iterations, are also set different values in the experiment,which demonstrates the feasibility of implementing GA with memristors.展开更多
基金the Key Research and Development Projects of Sichuan Science and Technology Department under Grant No.2018GZ0464the UESTC-ZHIXIAOJING Joint Research Center of Smart Home under Grant No.H04W210180.
文摘Activity recognition plays a key role in health management and security.Traditional approaches are based on vision or wearables,which only work under the line of sight(LOS)or require the targets to carry dedicated devices.As human bodies and their movements have influences on WiFi propagation,this paper proposes the recognition of human activities by analyzing the channel state information(CSI)from the WiFi physical layer.The method requires only the commodity:WiFi transmitters and receivers that can operate through a wall,under LOS and non-line of sight(NLOS),while the targets are not required to carry dedicated devices.After collecting CSI,the discrete wavelet transform is applied to reduce the noise,followed by outlier detection based on the local outlier factor to extract the activity segment.Activity recognition is fulfilled by using the bi-directional long short-term memory that takes the sequential features into consideration.Experiments in through-the-wall environments achieve recognition accuracy>95%for six common activities,such as standing up,squatting down,walking,running,jumping,and falling,outperforming existing work in this field.
基金supported by the Key Science and Technology R&D Project of Sichuan Province under Grants No. 2022YFG0038 and No. 2021YFG0018
文摘The knowledge graph(KG) that represents structural relations among entities has become an increasingly important research field for knowledge-driven artificial intelligence. In this survey, a comprehensive review of KG and KG reasoning is provided. It introduces an overview of KGs, including representation, storage, and essential technologies. Specifically, it summarizes several types of knowledge reasoning approaches, including logic rules-based, representation-based, and neural network-based methods. Moreover, this paper analyzes the representation methods of knowledge hypergraphs. To effectively model hyper-relational data and improve the performance of knowledge reasoning, a three-layer knowledge hypergraph model is proposed. Finally, it analyzes the advantages of three-layer knowledge hypergraphs through reasoning and update algorithms which could facilitate future research.
基金supported by the Research Fund of National Key Laboratory of Computer Architecture under Grant No.CARCH201501the Open Project Program of the State Key Laboratory of Mathematical Engineering and Advanced Computing under Grant No.2016A09
文摘In this paper, we approach the design of ID caching technology(IDCT) for graph databases, with the purpose of accelerating the queries on graph database data and avoiding redundant graph database query operations which will consume great computer resources. Traditional graph database caching technology(GDCT)needs a large memory to store data and has the problems of serious data consistency and low cache utilization. To address these issues, in the paper we propose a new technology which focuses on ID allocation mechanism and high-speed queries of ID on graph databases. Specifically, ID of the query result is cached in memory and data consistency is achieved through the real-time synchronization and cache memory adaptation. In addition, we set up complex queries and simple queries to satisfy all query requirements and design a mechanism of cache replacement based on query action time, query times, and memory capacity, thus improving the performance furthermore.Extensive experiments show the superiority of our techniques compared with the traditional query approach of graph databases.
基金supported by the Key Research and Development Projects of Sichuan Science and Technology Department under Grant No.2018GZ0464
文摘Indoor floor plans are of vital importance for a wide range of indoor social applications, however, they are often unavailable due to various reasons. This paper proposes a method to automatically construct the indoor floor plan with rooms and corridors as well as landmarks, using inertial traces collected with smartphones. Landmarks,such as turnings, doors, and stairs, are identified according to inertial sensors and WiFi signals. The inertial traces are then partitioned into segments according to the turnings and classified as room type or corridor type according to the doors. Clustering is applied on room type and corridor type trace segments separately to produce room type and corridor type clusters. The construction of room applies the α-shape algorithm on room type clusters and the construction of corridor employs the principal component analysis(PCA) algorithm or the α-shape algorithm on corridor type clusters. Evaluations in two representative scenarios show that the method can construct the floor plan of acceptable accuracy with a relatively small set of inertial traces.
文摘Several recent successes in deep learning(DL),such as state-of-the-art performance on several image classification benchmarks,have been achieved through the improved configuration.Hyperparameters(HPs)tuning is a key factor affecting the performance of machine learning(ML)algorithms.Various state-of-the-art DL models use different HPs in different ways for classification tasks on different datasets.This manuscript provides a brief overview of learning parameters and configuration techniques to show the benefits of using a large-scale handdrawn sketch dataset for classification problems.We analyzed the impact of different learning parameters and toplayer configurations with batch normalization(BN)and dropouts on the performance of the pre-trained visual geometry group 19(VGG-19).The analyzed learning parameters include different learning rates and momentum values of two different optimizers,such as stochastic gradient descent(SGD)and Adam.Our analysis demonstrates that using the SGD optimizer and learning parameters,such as small learning rates with high values of momentum,along with both BN and dropouts in top layers,has a good impact on the sketch image classification accuracy.
基金supported in part by the National Natural Science Foundation of China under Grant No.61503059
文摘Hyperparameters are important for machine learning algorithms since they directly control the behaviors of training algorithms and have a significant effect on the performance of machine learning models. Several techniques have been developed and successfully applied for certain application domains. However, this work demands professional knowledge and expert experience. And sometimes it has to resort to the brute-force search.Therefore, if an efficient hyperparameter optimization algorithm can be developed to optimize any given machine learning method, it will greatly improve the efficiency of machine learning. In this paper, we consider building the relationship between the performance of the machine learning models and their hyperparameters by Gaussian processes. In this way, the hyperparameter tuning problem can be abstracted as an optimization problem and Bayesian optimization is used to solve the problem. Bayesian optimization is based on the Bayesian theorem. It sets a prior over the optimization function and gathers the information from the previous sample to update the posterior of the optimization function. A utility function selects the next sample point to maximize the optimization function.Several experiments were conducted on standard test datasets. Experiment results show that the proposed method can find the best hyperparameters for the widely used machine learning models, such as the random forest algorithm and the neural networks, even multi-grained cascade forest under the consideration of time cost.
基金supported by the Research Fund of National Key Laboratory of Computer Architecture under Grant No.CARCH201501the Open Project Program of the State Key Laboratory of Mathematical Engineering and Advanced Computing under Grant No.2016A09
文摘In the era of Big Data, typical architecture of distributed real-time stream processing systems is the combination of Flume, Kafka, and Storm. As a kind of distributed message system, Kafka has the characteristics of horizontal scalability and high throughput, which is manly deployed in many areas in order to address the problem of speed mismatch between message producers and consumers. When using Kafka, we need to quickly receive data sent by producers. In addition, we need to send data to consumers quickly. Therefore, the performance of Kafka is of critical importance to the performance of the whole stream processing system. In this paper, we propose the improved design of real-time stream processing systems, and focus on improving the Kafka's data loading process.We use Kafka cat to transfer data from the source to Kafka topic directly, which can reduce the network transmission. We also utilize the memory file system to accelerate the process of data loading, which can address the bottleneck and performance problems caused by disk I/O. Extensive experiments are conducted to evaluate the performance, which show the superiority of our improved design.
基金the China Scholarship Council under Grant No.201406070059.
文摘Three-dimensional(3D)shape registration is a challenging problem,especially for shapes under non-rigid transformations.In this paper,a 3D non-rigid shape registration method is proposed,called balanced functional maps(BFM).The BFM algorithm generalizes the point-based correspondence to functions.By choosing the Laplace-Beltrami eigenfunctions as the function basis,the transformations between shapes can be represented by the functional map(FM)matrix.In addition,many constraints on shape registration,such as the feature descriptor,keypoint,and salient region correspondence,can be formulated linearly using the matrix.By bi-directionally searching for the nearest neighbors of points’indicator functions in the function space,the point-based correspondence can be derived from FMs.We conducted several experiments on the Topology and Orchestration Specification for Cloud Applications(TOSCA)dataset and the Shape Completion and Animation of People(SCAPE)dataset.Experimental results show that the proposed BFM algorithm is effective and has superior performance than the state-of-the-art methods on both datasets.
基金supported by the Research Fund for International Young Scientists of National Natural Science Foundation of China under Grant No. 61550110248the Sichuan Science and Technology Program under Grant No. 2019YFG0190
文摘This paper explores a way of deploying the classical algorithm named genetic algorithm(GA) with the memristor. The memristor is a type of circuit device with both characteristics of storage and computing, which provides the similarity between electronic devices and biological components, such as neurons, and the structure of the memristor-based array is similar to that of chromosomes in genetics. Besides, it provides the similarity to the image gray-value matrix that can be applied to image restoration with GA. Thus, memristor-based GA is proposed and the experiment about image restoration using memristor-based GA is carried out in this paper. And parameters,such as the size of initial population and the number of iterations, are also set different values in the experiment,which demonstrates the feasibility of implementing GA with memristors.
