CS786 Advanced Topics in Machine Learning

CS 786 - Spring 2013 - Advanced Topics in Machine Learning

News:

April 1: Project 1 due on Thursday April 4!
Feb 25: Project 1 data and sample solution file have been posted in the project webpage here Project 1.
Jan 25: class room changed to GITC4415
Jan 20: First class: Jan 24, 2013 @ 1pm in CKB 206

Class schedule: Thursdays 1:00 - 3:55 pm, CKB 206
Instructor: Zhi Wei ; Email: zhiwei@njit.edu; Office: GITC 3801
Office hours (GITC 3801): M 3-5pm, W 4:30-5:30pm, or by appointment.

Syllabus

Textbook (free!): The Elements of Statistical Learning: Data Mining, Inference, and Prediction (Second Edition)
　

Presentation schedule, Please download slides from Moodle.

Tentative Date	Lecture Topic	Presentation Leaders	Paper presentation	Readings
1/24	Introduction Slides Mixture Model Slides			Ch 8.5.1
1/31	Empirical Bayesian Model Slides	Nubyra	Pan et al. A mixture model approach to detecting differentially expressed genes with microarray data. Funct Integr Genomics 2003. presentation 1 (Nubyra)	Empirical Bayes
2/7	Markov Random Field model Slides	Xiang David	McLachlan et al. A simple implementation of a normal mixture approach to differential gene expression in multiclass microarrays. Bioinformatics 2006. Newton et al. On differential variability of expression ratios: improving statistical inference about gene expression changes from microarray data. J of Comp Bio 2001. presentation 1 (Xiang) presentation 2 (David)	MRFs
2/14	Feature and Model Selection Slides	Amiel	Kendziorski et al. On parametric empirical Bayes methods for comparing multiple groups using replicated gene expression profiles. Statist. Med. 2003. presentation 1 (Amiel)	Review
2/21	SVM Slides	Nafi	Wei and Li. A Markov Random Field Model for Network-based Analysis of Genomic Data. Bioinformatics 2007.
2/28	Linear model Slides	Chong Qingfeng	Zhang et al. Gene Selection Using Support Vector Machines With Nonconvex Penalty. Bioinformatics 2006. Wang et al. Hybrid huberized support vector machines for microarray classification and gene selection. Bioinformatics 2008. presentation 1 (Chong), presentation 2 (Qingfeng)	Fan and Li. Statistical challenges with high dimensionality: feature selection in knowledge discovery, Proceedings of the International Congress of Mathematicians, 2006. Fan and Lv. A selective overview of variable selection in high dimensional feature space. Statistica Sinica 2010.
3/7	Guest seminar/kernel learning	Dr. Pavel Kuksa	Multivariate and generalized similarity kernels for sequence analysis
3/14	Linear model with penalty Slides			Ch3.4, Ch18.3-18.4
3/21	Spring Break
3/28	Guest seminar/ultra-high dimension learning	Dr. Yang Feng	ISIS: A vehicle for the Universe of Sparsity
4/3	Guest seminar/large scale learning	Dr. Kai Zhang	Mining Big Data Using Kernel
4/4		Chao Qingfeng	Zhu and Hastie. Classification of gene microarrays by penalized logistic regression. Biostatistics 2004. Liu et al. Sparse Logistic Regression with Lp Penalty for Biomarker Identification. Stat Appl Genet Mol Biol 2007. presentation 1 (Chao), presentation 2 (Qingfeng)
4/11	Guest seminar/online learning	Dr. Slobodan Vucetic	Scalable Online Support Vector Machines
4/18	Dimension Reduction, Slides PPT, PDF Tree Extension, Slides PPT, PDF			PCA, Ch3.5.2, SIR Ch10, Ch15, Ch16
4/25	Guest seminar/Structured learning	Dr. Xinghua Lou	Structured Prediction with Reduced Annotations Tsochantaridis, Ioannis, et al. "Large margin methods for structured and interdependent output variables." Journal of Machine Learning Research 6.2 (2006): 1453. Teo, Choon Hui, et al. "Bundle methods for regularized risk minimization." The Journal of Machine Learning Research 11 (2010): 311-365. Lou, Xinghua and Hamprecht, Fred. "Structured learning from partial annotations." ICML, 2012.	Nowozin, Sebastian, and Christoph H. Lampert. Structured learning and prediction in computer vision. Vol. 6. Now publishers Inc, 2011.
5/2		Onur Kristin	Dai et al. Dimension Reduction for Classification with Gene Expression Microarray Data. Stat Appl Genet Mol Biol 2006 Shen and Tan. Dimension Reduction-Based Penalized Logistic Regression for Cancer Classification Using Microarray Data. IEEE/Acm Transactions On Comp Bio and Bioinform 2005 presentation 1 (Onur), presentation 2 (Kristin)

Academic integrityy

The NJIT Honor Code will be upheld, and any violations will be brought to the immediate attention of the Dean of Students.

Modifications to syllabus

The syllabus may be modified at the discretion of the instructor or in the event of extenuating circumstances.
Students will be notified in class of any changes to the syllabus.