模式识别与人工智能
Home      About Journal      Editorial Board      Instructions      Ethics Statement      Contact Us                   中文
Pattern Recognition and Artificial Intelligence
22 Judgement and Disposal of Academic Misconduct Article
22 Copyright Transfer Agreement
22 Proof of Confidentiality
22 Requirements for Electronic Version
More....
22 Chinese Association of Automation
22 National ResearchCenter for Intelligent Computing System
22 Institute of Intelligent Machines,Chinese Academy of Sciences
More....
 
 
2023 Vol.36 Issue.1, Published 2023-01-25
   
0
2023 Vol. 36 (1): 0-0 [Abstract] ( 336 ) [HTML 1KB] [ PDF 176KB] ( 335 )
1 Motion Planning under Uncertainty for Autonomous Driving:Opportunities and Challenges
ZHANG Xiaotong, WANG Jiacheng, HE Jingtao, CHEN Shitao, ZHENG Nanning

Motion planning algorithm, as an important part of autonomous driving systems, draws increasing attention from researchers. However, most existing motion planning algorithms only consider their application in deterministic structured environments, neglecting potential uncertainties in dynamic traffic environments. In this paper, motion planning algorithms are divided into two categories for the uncertain environment: partially observable Markov decision process and probability occupancy grid map.The two categories are introduced for three aspects: theoretical foundation, solution algorithm and practical application. The strategy with the maximum discounted reward in the future is calculated by partially observable Markov decision process based on the current confidence state. Probability occupancy grid map utilizes probability to represent the occupancy status of corresponding grids, measuring the possibility of dynamic changes in traffic flow, and well representing the uncertainty. Finally, the main challenges and future research directions for motion planning in uncertain environments are summarized .
2023 Vol. 36 (1): 1-21 [Abstract] ( 1055 ) [HTML 1KB] [ PDF 4027KB] ( 1574 )
22 A Concept Interaction-Based Cognitive Diagnosis Deep Model
ZHANG Suojuan, YU Xiaohan, CHEN Enhong, SHEN Shuanghong, ZHENG Yu, HUANG Song
Cognitive diagnosis is an intelligent assessment technique of mining learners' cognitive state based on learning data. Concepts in learning tasks are regarded as equally important by most cognitive diagnosis deep model. Without the consideration of the interaction between concepts, diagnosis accuracy is affected and interpretability is insufficient. To solve the problems, a concept interaction-based cognitive diagnosis deep model is proposed to realize the unified representation of students' cognitive state and concept weights. In the meanwhile, an algorithm of ideal response calculation based on the Choquet integral is implemented. Finally, a deep neural network based on fuzzy measures is proposed to predict learners' response performance. Experiments show that the proposed model holds advantages in prediction results and the explanation at the concept interaction level provided for prediction results.
2023 Vol. 36 (1): 22-33 [Abstract] ( 608 ) [HTML 1KB] [ PDF 1079KB] ( 647 )
34 Unsupervised Group Feature Selection Method for Graph Optimization Based on l2,0-norm Sparsity and Fuzzy Similarity
MENG Tiantian, ZHOU Shuisheng, TIAN Xinrun
Most graph-based unsupervised feature selection methods choose l2,1-norm sparse regularization of the projection matrix instead of non-convex l2,0-norm constraint. However, the l2,1-norm regularization method selects features one by one according to the scores, without considering the correlation of features. Therefore, an unsupervised group feature selection method for graph optimization based on l2,0-norm sparsity and fuzzy similarity is proposed, and it simultaneously performs graph learning and feature selection. In graph learning, the similarity matrix with exact connected components is learned. In the process of feature selection, the number of non-zero rows of projection matrix is constrained to realize group feature selection. To solve the non-convex l2,0-norm constraint, the feature selection vector with elements of 0 or 1 is introduced to transform the l2,0-norm constraint problem into 0-1 integer programming problem, and the discrete 0-1 integer constraint is transformed into two continuous constraints to solve the problem. Finally, fuzzy similarity factor is introduced to extend the method and learn more accurate graph structure. Experiments on real datasets show the effectiveness of the proposed method.
2023 Vol. 36 (1): 34-48 [Abstract] ( 487 ) [HTML 1KB] [ PDF 959KB] ( 968 )
49 Feature Subset Selection for Multi-scale Neighborhood Decision Information System
ZHANG Lujing, LIN Guoping, LIN Yidong, KOU Yi
Feature subset selection for multi-scale decision information system is an effective data preprocessing method for multi-scale classification problems. However, data types are often diverse and mixed in real application. The existing multi-scale models cannot handle these data effectively. To solve this problem, a formal definition of multi-scale neighborhood radius for multi-source heterogeneous multi-scale data is proposed in this paper. Multi-scale neighborhood information granule is constructed and its related properties are studied. Attribute significance is discussed, and a feature subset selection algorithm is proposed. Optimal scale selection and feature selection are conducted synchronously. By improving the Wu-Leung model, the scope of its application in practical problems is expanded to some extent. Finally, the feasibility and effectiveness of the proposed model and algorithm are verified on UCI datasets.
2023 Vol. 36 (1): 49-59 [Abstract] ( 473 ) [HTML 1KB] [ PDF 661KB] ( 592 )
60 Binary Acceleration and Compression for Dense Vector Entity Retrieval Models
WANG Yuanzheng, FAN Yixing, CHEN Wei, ZHANG Ruqing, GUO Jiafeng
In entity retrieval tasks, dense vector entity retrieval models are utilized to efficiently filter candidate entities related to a query from a large-scale entity base.However, the existing dense vector retrieval models engender low real-time computation efficiency and large required storage space due to the high dimension of entity vectors. In this paper,it is found that these entity vectors contain a large amount of redundant information through experiments. Most entity vectors are distributed in non-overlapping quadrants and quadrants containing entities with similar semantics are also closer to each other.Thus, a binary entity retrieval method is proposed to compress entity vectors and accelerate similarity calculations.Specifically, the sign function is employed to binary-compress high-dimensional dense floating-point vectors, and Hamming distance is exploited to speed up the retrieval.The reason that the proposed method can guarantee the retrieval performance is theoretically analyzed.The correctness of the theory is verified through qualitative and quantitative analysis experiments, and a method for improving binary retrieval performance based on random dimension increase and rotation is provided.
2023 Vol. 36 (1): 60-69 [Abstract] ( 367 ) [HTML 1KB] [ PDF 873KB] ( 737 )
70 Fusion Network Based on Progressive Nested Feature
SUN Junding, WANG Jinkai, TANG Chaosheng, WU Xiaosheng
In salient object detection, the computer detects the most interesting areas or objects in the visual scene by means of introducing the human visual attention mechanism. Aiming at the problems of unclear edge, incomplete object and missing detection of small objects in salient object detection, a fusion network based on progressive nested feature is proposed. Progressive compression module is adopted to continuously transfer and merge deeper features downward and make full use of advanced semantic information while the number of model parameters is reduced. A weighted feature fusion module is designed to aggregate the multi-scale features of the encoder into a feature map that can access both high-level and low-level information. Then, the aggregated features are allocated to other layers to fully obtain image context information and focus on small objects in the image. The asymmetric convolution block is introduced to further improve the detection accuracy. Experiments on six open datasets show that the proposed network achieves good detection results.
2023 Vol. 36 (1): 70-80 [Abstract] ( 378 ) [HTML 1KB] [ PDF 1166KB] ( 604 )
81 Policy Gradient Algorithm in Two-Player Zero-Sum Markov Games
LI Yongqiang, ZHOU Jian, FENG Yu, FENG Yuanjing
In two-player zero-sum Markov games, the traditional policy gradient theorem is only applied to alternate training of two players due to the influence of one player's policy on the other player's policy. To train two players at the same time, the policy gradient theorem in two-player zero-sum Markov games is proposed. Then, based on the policy gradient theorem, an extra-gradient based REINFORCE algorithm is proposed to achieve approximate Nash convergence of the joint policy of two players. The superiority of the proposed algorithm is analyzed in multiple dimensions. Firstly, the comparative experiments on simultaneous-move game show that the convergence and convergence speed of the proposed algorithm are better. Secondly, the characteristics of the joint policy obtained by the proposed algorithm are analyzed and these joint policies are verified to achieve approximate Nash equilibrium. Finally, the comparative experiments on simultaneous-move game with different difficulty levels show that the proposed algorithm holds a good convergence speed at higher difficulty levels.
2023 Vol. 36 (1): 81-91 [Abstract] ( 391 ) [HTML 1KB] [ PDF 884KB] ( 688 )
92
2023 Vol. 36 (1): 92-94 [Abstract] ( 196 ) [HTML 1KB] [ PDF 219KB] ( 270 )
模式识别与人工智能
 

Supervised by
China Association for Science and Technology
Sponsored by
Chinese Association of Automation
NationalResearchCenter for Intelligent Computing System
Institute of Intelligent Machines, Chinese Academy of Sciences
Published by
Science Press
 
Copyright © 2010 Editorial Office of Pattern Recognition and Artificial Intelligence
Address: No.350 Shushanhu Road, Hefei, Anhui Province, P.R. China Tel: 0551-65591176 Fax:0551-65591176 Email: bjb@iim.ac.cn
Supported by Beijing Magtech 　Email:support@magtech.com.cn