γ-EM approach to latent orientations for cryo-electron microscopy image clustering analysis

以作者查詢圖書館館藏

、以作者查詢臺灣博碩士

、以作者查詢全國書目

、勘誤回報

、線上人數：25

、訪客IP：18.224.93.225

姓名

林哲宇(Che-YU Lin) 查詢紙本館藏

畢業系所

統計研究所

論文名稱

(γ-EM approach to latent orientations for cryo-electron microscopy image clustering analysis)

相關論文

★ 長期追蹤資料上的 Gamma-EM 分群	★ Contrastive Principal Component Analysis for High Dimension, Low Sample Size Data
★ Bayesian method for sparse principal component analysis	★ Sparse Bayesian Estimation with High-dimensional Binary Response Data
★ Q學習結合監督式學習在股票市場的應用	★ 隱私差分應用於聯邦式水平分割主成分分析
★ Contrastive Principal Component Analysis for High-Dimension, Low-Sample-Size Data with Noise-Reduction	★ γ-SUP 演算法在 NBA 資料分析上的應用
★ 基於Q-learning與非監督式學習之交易策略	★ 視覺化股票市場之狀態變動
★ Principal Components on t-SNE	★ gamma-SUP on PCA
★ 高維度環境下Kronecker包絡主成分分析的漸近性

檔案

[Endnote RIS 格式]

[Bibtex 格式]

[相關文章]

[文章引用]

[完整記錄]

[館藏目錄]

[檢視]

[下載]

本電子論文使用權限為同意立即開放。
已達開放權限電子全文僅授權使用者為學術研究之目的，進行個人非營利性質之檢索、閱讀、列印。
請遵守中華民國著作權法之相關規定，切勿任意重製、散佈、改作、轉貼、播送，以免觸法。

摘要(中)

影像分類分析廣泛應用於許多科學領域，例如計算機視覺、異常檢測、生物學、醫學等。基於最小化 Kullback-Leibler (KL) 散度估計的最大概似估計方法在影像分類中很流行。然而，眾所周知，最大概似估計值對模型假設很敏感。如果存在數據異常值或模型錯誤設定，則可能導致有偏差的估計和不穩定的推論。其中一種補救措施是利用穩健的散度。在本篇論文中，我們引入了更一般化的 KL 散度，稱為 γ-散度。它提供了一種穩健的方法來排除異常值和模型錯誤指定的問題。我們提出了一個基於局部最小化 γ-散度的半參數分群結構，進一步考慮了影像的旋轉和平移性質。我們將所提出的方法命名為 γ-最大化期望值演算法。γ-最大化期望值演算法可以改進參數的估計。模擬研究結果顯示，與現今流行的方法相比，本篇所提出的方法表現良好。

摘要(英)

Image clustering analysis is widely used in many scientific fields such as computer vision, anomaly detection, biology, medicine, etc. Maximum-likelihood based methods, which arise from the minimum Kullback-Leibler (KL) divergence estimation, are popular in image clustering. However, MLE is known to be sensitive to model assumptions. It can lead to biased estimation and unstable inference if there exist data outliers or model misspecification. A remedy is to introduce a robust divergence. In this thesis, a generalization of KL divergence, namely γ-divergence, is brought in. It provides a robust method to defend outliers and model misspecification. We propose a semi-parametric framework for clustering based on local minimization of γ-divergence, further considering rotation and translation of images. We name the proposed method the γ-Expectation–Maximization (γ-EM) algorithm. The γ-EM leads to improved parameter estimation. Numerical studies have demonstrated that the proposed method performs reasonably well in comparison with some popular methods.

關鍵字(中)

★ 分群演算法
★ 最大化期望值演算法
★ γ-散度
★ 最大概似估計法

關鍵字(英)

★ clustering algorithm
★ EM-algorithm
★ γ-divergence
★ maximum-likelihood method

論文目次

摘要 i
Abstract ii
致謝 iii
List of Figures vi
1 Introduction 1
1.1 Expectation-Maximization (EM) Algorithm . ................ 3
1.2 γ-Divergence . ................................ 4
1.3 q-Gaussian Distribution . ........................... 5
1.4 γ-SUP . .................................... 6
2 Method 9
2.1 γ-EM . .................................... 9
2.2 Accelerated Approach . ........................... 13
2.2.1 Average pooling Method . ...................... 13
2.2.2 Random Partition . .......................... 14
2.2.3 Substituting θ(⋆) for θ(t) . ...................... 14
v
3 Theory 17
3.1 γ-Likelihood . ................................ 17
3.2 Proof of Theorem . .............................. 20
4 Numerical Study 23
4.1 Density-Based Spatial Clustering of Applications with Noise (DBSCAN) . 24
4.2 Simulatedimages . .............................. 25
4.2.1 Case1 : clustering with SNR=0.1 . ................ 26
4.2.2 Case2 : clustering with SNR=0.05 . ................ 27
4.3 Compare γ-SUP and γ-EM . ......................... 30
5 Application : synthetic cryo-EM data 33
5.1 Clustering with SNR=0.2 and 0.1 . ..................... 34
5.2 Compare γ-EM and γ-SUP . ......................... 36
6 Conclusion 40
Bibliography 41

參考文獻

Chen, T.-L.,Hsieh,D.-N.,Hung,H.,Tu,I.-P.,Wu,P.-S.,Wu,Y.-M.,…others(2014). γ-sup: Aclusteringalgorithmforcryo-electron microscopy images of asymmetric particles. Annals of Applied Statistics, 8(1), 259–285.
Dempster,A.P.,Laird,N.M.,&Rubin,D.B (1977).Maximum likelihood from incomplete data via the em algorithm. Journal of the Royal Statistical Society: Series B
(Methodological), 39(1), 1–22.
Eguchi, S.,Komori,O.,&Kato,S.(2011).Projective power entropy and maximum tsallis entropy distributions. Entropy, 13(10), 1746–1764.
Ester,M.,Kriegel,H.-P.,Sander,J.,Xu,X.,etal.(1996).Adensity-based algorithm for
discovering clusters in large spatial databases with noise.In Kdd (Vol.96,pp.226–231).
Fujisawa, H.,&Eguchi,S.(2008).Robust parameter estimation with a small bias against heavy contamination. Journal of Multivariate Analysis, 99(9), 2053–2081.
Hung, H.,Jou,Z.-Y.,&Huang,S.-Y.(2018).Robust mislabel logistic regression without modeling mislabel probabilities. Biometrics, 74(1), 145–154.
Jones, M.,Hjort,N.L.,Harris,I.R.,&Basu,A.(2001).A comparison of related density-based minimum divergence estimators. Biometrika, 88(3), 865–873.
Kawashima, T.,&Fujisawa,H (2017).Robust and sparse regression via γ-divergence. Entropy, 19(11),608.
Nakagawa, T.,&Hashimoto,S.(2020).Robust bayesian inference via γ-divergence. Communications in Statistics-Theory and Methods, 49(2), 343–360.
Notsu, A.,Komori,O.,&Eguchi,S.(2014).Spontaneous clustering via minimum gamma-divergence. Neural computation, 26(2), 421–448.
Punjani, A.,Rubinstein,J.L.,Fleet,D.J.,&Brubaker,M.A.(2017).cryosparc:algorithms for rapid unsupervised cryo-em structure determination. Nature methods, 14(3), 290–
296.
Scheres, S.H (2021).Relion:implementation of a bayesian approach to cryo-em structure determination. Journal of Structural Biology, 180, 519–530.
Shiu, S.,&Chen,T.(2012).Clustering by self-updating process. arXiv preprint arxiv:1201.1979.

指導教授

王紹宣(Shao-Hsuan Wang)

審核日期

2021-7-15

推文