Unsupervised pseudoinverse hashing learning model for rare astronomical object retrieval

作     者：WANG Ke GUO Ping LUO ALi XU MingLiang WANG Ke;GUO Ping;LUO ALi;XU MingLiang

作者机构：Artificial Intelligence Research CenterSchool of Computer and Artificial IntelligenceZhengzhou UniversityZhengzhou 450001China School of Systems ScienceBeijing Normal UniversityBeijing 100875China Key Laboratory of Optical AstronomyNational Astronomical ObservatoriesChinese Academy of SciencesBeijing 100012China 

出 版 物：《Science China(Technological Sciences)》 (中国科学（技术科学英文版）)

年 卷 期：2022年第65卷第6期

页      面：1338-1348页

核心收录：

学科分类：07[理学] 08[工学] 081203[工学-计算机应用技术] 070401[理学-天体物理] 0805[工学-材料科学与工程（可授工学、理学学位）] 0704[理学-天文学] 0812[工学-计算机科学与技术（可授工学、理学学位）] 

基　　金：supported by the Postdoctoral Science Foundation of China(Grant No.2020M682348) the Key Research Foundation of Henan Higher Education Institutions(Grant No.21A520002) the National Key Research and Development Program of China(Grant No.2018AAA0100203) the Joint Research Fund in Astronomy(Grant No.U1531242)under a cooperative agreement between the National Natural Science Foundation of China and the Chinese Academy of Sciences(CAS)。 

主　　题：compact features unsupervised hashing object retrieval pseudoinverse learning 

摘      要：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 spectra retrieval.The proposed method employs a multilayer neural network using autoencoders as the local compact feature extractors.Autoencoders are trained with a non-gradient learning algorithm with graph Laplace regularization.This algorithm also simplifies the tuning of network architecture hyperparameters and the learning control hyperparameters.Meanwhile,the graph Laplace regularization can enhance the robustness by reducing the sensibility to noise.The proposed model is data-driven;thus,it can be viewed as a general-purpose retrieval model.The proposed model is evaluated in experiments and real-world applications where rare Otype stars and their subclass are retrieved from the dataset obtained from the Large Sky Area Multi-Object Fiber Spectroscopic Telescope(Guo Shoujing Telescope).The experimental and application results show that the proposed model outperformed the baseline methods,demonstrating the effectiveness of the proposed method in rare spectra retrieval tasks.