공분산을 활용한 콘크리트 오토인코더 기반 비지도 특징 선택 기법 연구 = CoCoder : Concrete Autoencoder using Covariance for Unsupervised Feature Selection

국문 초록 (Abstract)

특징 선택은 특징 공학의 한 과정으로 주어진 정형 데이터로부터 유의미한 특징 (feature, column)을 선택하는 것을 목적으로 한다.
딥러닝 기술이 다양한 분야에서 주목할 만한 수행 능력을 보여줌에 따라 특징 선택 분야에서도 딥러닝 기술 기반 연구가 활발히 이루어지고 있다. 본 논문에서는 concrete autoencoder 기반 선택 기법에 주목하였다. Concrete autoencoder란 autoencoder에 concreterandom variable을 적용하여 유의미한 특징을 선택하는 기법이다. 하지만 concrete autoencoder 기법은 특징 선택 시 중복을 허용하고,저차원 벡터 공간 내에서 데이터가 클래스별로 군집화 되지 않는다는 문제가 있다. 따라서 본 논문은 저차원 벡터 공간 내에서 데이터의 특징별 covariance를 고려하는 기법을 제시하고 다양한 데이터를 사용하여 이 기법을 평가한다. 제안하는 방법은 특히 유전적 정보를 담고 있는 바이오 데이터를 사용했을 때 우수한 성능을 보여준다.

다국어 초록 (Multilingual Abstract)

Feature selection is a feature engineering process that aims to select meaningful features from given structured data. As deeplearning technology shows remarkable performance in various fields, deep learning-based research is also actively studied in thefeature selection field. In particular, the concrete autoencoder method, which selects important features by applying a concreterandom variable to the autoencoder, presented excellent performance in the field of feature selection. However, the concreteautoencoder allows overlap when selecting features and has the problem that data is not clustered by class within alow-dimensional vector space. In this paper, we propose a new feature selection technique that is based on the concreteautoencoder technique and can compensate for the shortcomings of the concrete autoencoder technique. The proposed methodconsiders the covariance of data features in a low-dimensional domain to prevent the redundant feature selection while improvingthe clustering quality of samples within a low-dimensional space. The proposed technique showed superior feature selectionperformance compared to existing techniques, and its superiority is especially evident for biological data containing geneticinformation.

참고문헌 (Reference)

• 1 Battiti R, "Using Mutual Information for Selecting Features in Supervised Neural Net Learning" 5 (5): 537-550, 1994
• 2 Mahsa Samareh Jahani, "Unsupervised feature selection guided by orthogonal representation of feature space" 516 : 61-76, 2023
• 3 Deng Cai, "Unsupervised feature selection for multi-cluster data" 333-342, 2010
• 4 Renato Cordeiro de Amorim, "Unsupervised feature selection for large data sets" 128 : 183-189, 2019
• 5 Pengfei Zhu, "Unsupervised feature selection by regularized selfrepresentation" 48 (48): 438-446, 2015
• 6 Pengfei Zhu, "Unsupervised feature selection by regularized selfrepresentation" 28 (28): 438-446, 2015
• 7 Maddison, C. J., "The concrete distribution : A continuous relaxation of discrete random variables"
• 8 Samuel H. Huang, "Supervised feature selection : A tutorial" 4 (4): 22-37, 2015
• 9 Gumbel, Emil Julius, "Statistical theory of extreme values and some practical applications a series of lectures, Vol.33" US Government Printing Office
• 10 Petricoin EF 3rd, "Serum proteomic patterns for detection of prostate cancer" 94 (94): 1576-1578, 2002
• 11 Robert Tibshirani, "Regression Shrinkage and Selection Via the Lasso" 58 (58): 267-288, 1996
• 12 Sun, Lixin, "Neuronal and glioma-derived stem cell factor induces angiogenesis within the brain" 9 : 287-300, 2006
• 13 Xiaofei He, "Laplacian Score for Feature Selection" 18 : 2005
• 14 Cai Deng, "Graph regularized nonnegative matrix factorization for data representation" 33 (33): 1548-1560, 2011
• 15 Y. Lecun, "Gradient-based learning applied to document recognition" 86 (86): 2278-2324, 1998
• 16 X. Liu, "Global and Local Structure Preservation for Feature Selection" 25 (25): 1083-1095, 2014
• 17 Holland, John H., "Genetic Algorithms" 267 (267): 66-73, 1992
• 18 William A. Freije, "Gene Expression Profiling of Gliomas Strongly Predicts Survival" 64 (64): 6503-6510, 2004
• 19 Han Xiao, "Fashion-MNIST: a Novel Image Dataset for Benchmarking Machine Learning"
• 20 G. Katz, "ExploreKit : Automatic Feature Generation and Selection" 979-984, 2016
• 21 Wang, S., "Embedded Unsupervised Feature Selection" 29 (29): 2015
• 22 Xu Chu, "Data Cleaning : Overview and Emerging Challenges" 2201-2206, 2016
• 23 Muhammed Faith Balin, "Concrete Autoencoders: Differentiable Feature Selection and Reconstruction" 97 : 444-453, 2019
• 24 U. Alon, "Broad patterns of gene expression revealed by clusterꠓing analysis of tumor and normal colon tissues probed by oligonucleotide arrays" 96 (96): 6745-6750, 1999
• 25 Hinton, Geoffrey E, "Autoencoders, Minimum Description Length and Helmholtz Free Energy" 6 : 1993
• 26 K. Han, "Autoencoder Inspired Unsupervised Feature Selection" 2941-2945, 2018
• 27 Xinxing Wu, "Algorithmic stability and generalization of an unsupervised feature selection algorithm" 34 : 19860-19875, 2021
• 28 Amir Moslemi, "A tutorial-based survey on feature selection : Recent advancements on feature selection" 126 : 107136-, 2023
• 29 Razieh Sheikhpour, "A Survey on semi-supervised feature selection methods" 64 : 141-158, 2017
• 30 Narendra, Patrenahalli M., "A Branch and Bound Algorithm for Feature Subset Selection" C-26 (C-26): 917-922, 1977