BLG 601E – Pattern Recognition, Spring 2015

Department of Computer Engineering, Istanbul Technical University

Instructor: Zehra Cataltepe        Schedule: Wed. 9:30-12:30 (Room: 5306)

Please visit www.ninova.itu.edu.tr for slides, to submit your homeworks and other current information about the class.

Course Goals

Gain knowledge and practical experience in Bayesian pattern recognition.

Gain knowledge about latest developments and applications in (Bayesian) pattern recognition.

Learn practical skills and analytic background for building and enhancing pattern recognition applications.

Prerequisites                                                                                               

BLG527E (Machine Learning)

Probability, statistics, linear algebra.

Ability to use at least one of the following languages: MATLAB, R, python, java.

Understanding of different Pattern Recognition concepts and algorithms.

If you have no previous Pattern Recognition/Machine Learning experience, then please first take the MS level class BLG527E (Machine Learning).

In BLG527E (Machine Learning), we learn the basics of machine learning/pattern recognition from a frequentist perspective. In BLG601E we cover similar topics but there is added emphasis on Bayesian approaches to different machine learning techniques and also students present some recent research papers on each topic.

 

Grading :

Project

1

30%

Homeworks

6

(2 HWs will be paper presentations)

30%

Final

You have to have at least 30% over the homeworks and project to enter the final exam.

1

40%

 

 

TentativeWeekly Program

 

Week

Date

Content

1

Feb 4

Introduction, mathematical preliminaries

Pattern Recognition basics [CB1]

2

Feb 11

Probability Distributions [CB2]

3

Feb 18

No class due to snow!

4

Feb 25

Probability Distributions [CB2]

5

Mar 4

Linear Models for Regression [CB3] (HW1) (paperlist Announced)

Linear Models for Classification [CB4]

6

Mar  11

Linear Models for Classification (contd) [CB4]

Graphical Models  [CB8] 

7

Mar 18

 

Graphical Models  [CB8]

6.4. Gaussian Processes

7.2. Relevance Vector Machines

 (Project and paper selection)

8

Mar 25

Basics of: Deep Neural Networks, Bayesian Learning, Online Learning, Big Data Learning

(HW2) Paper and project discussion

9

Apr 1

Mixture Models and EM [CB9]

10

Apr 8

Approximate Inference [CB10]

(HW3)

11

Apr 15

Sampling Methods [CB11]

(HW4) Project Preliminary Reports Due

12

Apr 22

Current Topics (Deep Learning, BigData, Social Networks, Bayesian Learning)

Paper Presentations/Project discussions

 

Apr 29

Current Topics (Deep Learning, BigData, Social Networks, Bayesian Learning)

Paper Presentations/Project discussions

(HW5&6: Paper Reports due)

13

May 6

Project Presentations and Reports

 

 

Paper Presentation Schedule:

 

Who

Week

Paper

Cemal, Pınar

Apr-22

Kahou, S. E., Pal, C., Bouthillier, X., Froumenty, P., Gülçehre, Ç., Memisevic, R., ... & Wu, Z. (2013, December). Combining modality specific deep neural networks for emotion recognition in video. In Proceedings of the 15th ACM on International conference on multimodal interaction (pp. 543-550). ACM.

Göksu, Nazanin

Apr-22

Kimovski, D., Ortega, J., Ortiz, A., & Baños, R. (2015). Parallel alternatives for evolutionary multi-objective optimization in unsupervised feature selection. Expert Systems with Applications, 42(9), 4239-4252.

İsmail, Cumali, Beyza

Apr-22

Chapelle, O., & Zhang, Y. (2009, April). A dynamic bayesian network click model for web search ranking. In Proceedings of the 18th international conference on World wide web (pp. 1-10). ACM.

İsmail, Cumali, Beyza

Apr-22

Sutskever, I., Vinyals, O., & Le, Q. V. (2014). Sequence to sequence learning with neural networks. In Advances in Neural Information Processing Systems (pp. 3104-3112).

Sacide, Ozan

Apr-22

Dahl, George E., Adams, Ryan P., and Larochelle, Hugo. Training Restricted Boltzmann Machines on word observations. In International Conference on Machine Learning, 2012.

Kübra, Şirin

Apr-22

Benson, A. R., Lee, J. D., Rajwa, B., & Gleich, D. F. (2014). Scalable methods for nonnegative matrix factorizations of near-separable tall-and-skinny matrices. In Advances in Neural Information Processing Systems (pp. 945-953).

Barış, Yiğit

Apr-22

Yang Mu, Wei Ding, Tianyi Zhou, and Dacheng Tao. 2013. Constrained stochastic gradient descent for large-scale least squares problem. In Proceedings of the 19th ACM SIGKDD international conference on Knowledge discovery and data mining (KDD '13), Inderjit S. Dhillon, Yehuda Koren, Rayid Ghani, Ted E. Senator, Paul Bradley, Rajesh Parekh, Jingrui He, Robert L. Grossman, and Ramasamy Uthurusamy (Eds.). ACM, New York, NY, USA, 883-891. DOI=10.1145/2487575.2487635 http://doi.acm.org/10.1145/2487575.2487635

Majid, Pembe

Apr-22

Paisley, J., Blei, D., & Jordan, M. I. (2015). Bayesian nonnegative matrix factorization with stochastic variational inference. Handbook of Mixed Membership Models and Their Applications. Chapman and Hall/CRC.

Cemal, Pınar

Apr-29

 Zhu, Z., Luo, P., Wang, X., & Tang, X. (2014). Multi-view perceptron: a deep model for learning face identity and view representations. In Advances in Neural Information Processing Systems (pp. 217-225).

Göksu, Nazanin

Apr-29

Halawi, G., Dror, G., Gabrilovich, E., & Koren, Y. (2012, August). Large-scale learning of word relatedness with constraints. In Proceedings of the 18th ACM SIGKDD international conference on Knowledge discovery and data mining (pp. 1406-1414). ACM.

İsmail, Cumali, Beyza

Apr-29

del Río, S., López, V., Benítez, J. M., & Herrera, F. (2014). On the use of MapReduce for imbalanced big data using Random Forest. Information Sciences, 285, 112-137.

Sacide, Ozan

Apr-29

Ciresan, Dan, Ueli Meier, and Jürgen Schmidhuber. "Multi-column deep neural networks for image classification." Computer Vision and Pattern Recognition (CVPR), 2012 IEEE Conference o". IEEE, 2012."

Kübra, Şirin

Apr-29

Broderick, T., Boyd, N., Wibisono, A., Wilson, A. C., & Jordan, M. I. (2013). Streaming variational bayes. In Advances in Neural Information Processing Systems (pp. 1727-1735).

Barış, Yiğit

Apr-29

Graepel, T., Candela, J. Q., Borchert, T., & Herbrich, R. (2010). Web-scale bayesian click-through rate prediction for sponsored search advertising in microsoft's bing search engine. In Proceedings of the 27th International Conference on Machine Learning (ICML-10) (pp. 13-20).

Majid, Pembe

Apr-29

Xiao, L. (2009). Dual averaging method for regularized stochastic learning and online optimization. In Advances in Neural Information Processing Systems (pp. 2116-2124).

 

 

Project:

Choose a groupmate.  Possible project topic(s) will be announced on March 4.

5% preliminary report

25% project report

 

 

References:

by Zehra Cataltepe, Last update: Mar 24,  2015.