Dr. Murat Dundar
Associate Professor
Computer & Information Science Dept.

   

Contact Information

Office Address:
Computer & Information Science Dept.
723 W. Michigan St., SL 280C, IUPUI
Indianapolis, IN 46202

Tel: 317-278-6488
Fax: 317-274-9742

E-mail:
mdundar 'at' iupui.edu

 

 

SHORT BIO:

I received my BS degree from Bogazici University, Istanbul, Turkey, in 1997 and MS and PhD degrees from Purdue University, West Lafayette, IN, USA, in 1999 and 2003 respectively, all in Electrical Engineering. Between 2003 and 2008 I was with the CAD and Knowledge Solutions group of Siemens Health. At Siemens Health, I was involved in the development of a broad spectrum of computer aided diagnosis/detection applications including FDA-approved Lung and Colon CAD products. I have joined IUPUI Computer and Information Science Department as a tenure-track assistant professor in 2008, where I became an associate professor in 2014.

RESEARCH INTERESTS:

My area of expertise is in machine learning with a special focus on open-world machine learning, where we aim to replace the traditional brute-force approach of fitting a fixed model onto the data with more flexible models that can account for the non-stationary nature of real-world machine learning problems. We achieve this by placing suitably chosen non-parametric Bayesian priors over class distributions to model not only observed classes but unobserved ones as well in an effort to perform joint classification and clustering. Deep feature learning has recently become an important component of my research in open-world machine learning. How to learn a feature embedding that is specific enough to discriminate observed classes but also general enough to accommodate a large number of unobserved classes? How to develop loss functions that are not biased towards observed classes? How to take advantage of class descriptions or other side information to discover and model unobserved classes? These are some of the deep machine learning problems I have been working lately. My most recent research projects involve axon segmentation in electron microscopy images, mineral mapping on Mars using orbital data, describing the undescribed species using images and DNA, authorship attribution in historical texts. For more information on my research please visit my project page or Google Scholar research profile.

SELECT PAPERS:

Emanuele Plebani, Bethany L. Ehlmann, Ellen K. Leask, Valerie K. Fox, M. Murat Dundar, "A Machine Learning Toolkit for CRISM Image Analysis," Icarus, January 2022. Online

Emanuele Plebani, Natalia P. Biscola, Leif A. Havton, Bartek Rajwa, Abida Sanjana Shemonti, Deborah Jaffey, Terry Powley, Janet R. Keast, Kun-Han Lu, M. Murat Dundar, "High-throughput segmentation of unmyelinated axons by deep learning," Scientific Reports 12, 1198, 2022. Online

Sarkhan Badirli, Zeynep Akata, George Mohler, Christine Picard, Murat Dundar, "Fine-Grained Zero-Shot Learning with DNA as Side Information," In Advances on Neural Information Processing Systems (NeurIPS'21), 2021. Online

Sarkhan Badirli, Mary Borgo Tan, Abdulmecit Gungor, Murat Dundar, "Open Set Authorship Attribution toward Demystifying Victorian Periodicals," In International Conference on Document Analysis and Recognition, pp. 221-235. Springer, Cham, 2021. Online  

Sarkhan Badirli, Zeynep Akata, Murat Dundar, "Bayesian Zero-shot Learning," ECCV Workshops, 2020. Online

Murat Dundar, Bethany Ehlmann, Ellen Leask, "Machine-Learning-Driven New Geologic Discoveries at Mars Rover Landing Sites: Jezero and NE Syrtis," Earth and Space Science Open Archive, 2019. PDF 

Yicheng Cheng, Bartek Rajwa, Murat Dundar, "Bayesian Nonparametrics for Non-exhaustive Learning," Advances in Neural Information Processing Systems (NIPS), Bayesian Nonparametrics Workshop, 2018.  Online

Yicheng Cheng, Murat Dundar, George Mohler, "A coupled ETAS-I2GMM point process with applications to seismic fault detection," The Annals of Applied Statistics, Volume 12 (3), pp. 1853-1870, 2018. Online

Halid Z. Yerebakan and Murat Dundar, “Partially Collapsed Parallel Gibbs Sampler for Dirichlet Process Mixture Models,” Pattern Recognition Letters, Volume 90, Pages 22-27, 2017.

Murat Dundar, Qiang Kou, Baichuan Zhang, Yicheng He, Bartek Rajwa, “Simplicity of Kmeans versus Deepness of Deep Learning: A Case of Unsupervised Feature Learning with Limited Data,” In Proceedings of IEEE International Conference on Machine Learning Applications, Miami, FL, USA, December 11-13, 2015. PDF

Halid Z. Yerebakan, Bartek Rajwa, Murat Dundar, "The Infinite Mixture of Infinite Gaussian Mixtures," Advances in Neural Information Processing Systems (NIPS), 2014.  PDF

Murat Dundar, Ferit Akova, Halid Z. Yerebakan, Bartek Rajwa, "A Non-parametric Bayesian Model for Joint Cell Clustering and Cluster Matching: Identification of Anomalous Sample Phenotypes with Random Effects," BMC Bioinformatics 15 (1), 314, 2014. Online

Murat Dundar, Halid Z. Yerebakan, Bartek Rajwa, "Batch Discovery of Recurring Rare Classes toward Identifying Anomalous Samples," In Proceedings of the 20th Annual SIGKDD International Conference on Knowledge Discovery and Data Mining (SIGKDD'14), New York, USA, Aug 24-27 2014.  PDF Video Lecture

Murat Dundar, Ferit Akova, Yuan Qi, Bartek Rajwa, “Bayesian Nonexhaustive Learning for Online Discovery and Modeling of Emerging Classes,” In John Langford and Joelle Pineau (Eds.), Proceedings of the 29th International Conference on Machine Learning (ICML'12), Edinburgh, Scotland, June 26-July 1, 2012 (pp. 113-120). Omnipress, 2012. PDF

Murat Dundar, Sunil Badve, Gokhan Bilgin, Vikas Raykar, Olcay Sertel, Metin N. Gurcan, “Computerized Classification of Intraductal Breast Lesions using Histopathological Images”, IEEE Transactions on Biomedical Engineering, 58(7):1977-1984, 2011. PDF

Murat Dundar, Matthias Wolf, Sarang Lakare, Marcos Salganicoff, Vikas Raykar “Polyhedral Classifiers for Target Detection: A Case Study: Colorectal Cancer”, In Proceedings of the 25th International Conference on Machine Learning (ICML 2008), pp.288-295, Helsinki, July 2008. PDF

Murat Dundar, Glenn Fung, Balaji Krishnapuram, Bharat Rao, “Multiple Instance Learning Algorithms for Computer Aided Diagnosis”, IEEE Transactions on Biomedical Engineering, Volume 55, No. 3, pp 1005-1015, March 2008. PDF

Murat Dundar, Jinbo Bi, “Joint optimization of cascaded classifiers for computer aided detection”, In Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR ‘07), Minneapolis, Minnesota, USA, 18-23 June 2007. IEEE Computer Society 2007. (Oral paper, acceptance rate: 5%) PDF

Murat Dundar, Balaji Krishnapuram, Jinbo Bi, Bharat Rao, “Learning from Non-IID Data”, In Proceedings of the 20th International Joint Conference on Artificial Intelligence (IJCAI 2007), Hyderabad, India, January 6-12, 2007. PDF

Glenn Fung, Murat Dundar, Balaji Krishnapuram, Bharat Rao, “Multiple Instance Algorithms for Computer Aided Diagnosis”, Advances in Neural Information Processing Systems 19 (NIPS 2006), Vancouver, CA, 2006. PDF

Murat Dundar, Glenn Fung, Jinbo Bi, Sandilya Sathyakama, Bharat Rao, “Sparse Fisher Discriminant Analysis for Computer Aided Detection”, In Proceedings of the SIAM International Conference of Data Mining (SDM ’05), Newport Beach, CA, USA, 2005. PDF

FUTURE STUDENTS:

I am actively recruiting graduate students who have keen interest in deep machine learning and their applications to problems in life sciences. The IUPUI doctoral/master programs in computer sciences awards Purdue University degrees.

Background: An ideal candidate should have a BS in CS/ECE/Math/Stats and have a strong background in linear algebra, calculus, probability and random variables.

Must-have skills:

  • Extensive experience with Matlab/Python/C programming.
  • Good English writing and presentation skills
  • Ability to do independent research

NEWS:

Unmyelinated Fiber Segmentation (6/21)
First published approach studying automated segmentation of unmyelinated fibers in transmission electron microscopy images.

PhD Thesis Defense (12/21)
Congratulations to Sarkhan Badirli for successfully defending his PhD Thesis entitled "Fine Grained Zero-Shot Object Recognition".

PhD Thesis Defense (7/21)
Congratulations to Yicheng Cheng for successfully defending his PhD Thesis entitled "Machine Learning in the Open World".

CRISM Machine Learning Toolkit (6/21)
Machine learning toolkit that implements our hierarchical Bayesian classifier for CRISM image analysis.

NASA Grant Awarded (7/19)
Grant with Bethany Ehlmann “A Quantitative Approach to Understanding the Distribution and Diversity of Key Water-Formed Minerals on Mars” was awarded by NASA MDAP program.

NIH Grant Awarded (11/18)
Grant with Ji-Xin Cheng “Quantitative SRS Imaging of Cancer Metabolism at Single Cell Level” was awarded by NIH/NCI.

New Study Raises Questions About Salts Near Seasonally Darkening Streaks on Mars (11/18)
Our non-parametric Bayesian algorithm helped us detect a previously unknown artifact in CRISM data.

MS Thesis Defense (4/18)
Congratulations to Abdulmecit Gungor for successfully defending his MS Thesis entitled "Benchmarking Authorship Attribution Techniques"

Roche Global Code4Life Competition (3/18)
Congratulations to Abdulmecit Gungor, Sarkhan Badirli, and Sarun Gulyanon for their outstanding achievement in the Roche Global Competition.
IUPUI takes second place in Roche Global Code4life University Challenge

PhD Thesis Defense (5/17)
Congratulations to Halid Yerebakan for successfully defending his PhD Thesis entitled "Hierarchical Non-Parametric Bayesian Mixture Models and Applications on Big Data".

Computer trained to predict which AML patients will go into remission, which will relapse (2/17)
Highlighting our work on doubly non-parametric Bayesian clustering and its application to AML remission/relapse prediction

Training computers to differentiate between people with the same name (1/17)
Highlighting our work on name disambiguation

Undergraduate Research Award (3/15)
Congratulations to Nhan Do for receiving IUPUI Undergraduate Research Program (RISE) Award.

New Software (5/14)
Group clustering with random effects implemented in C++ for clustering and cluster matching across a batch of samples. Download Here.

PhD Thesis Defense (6/13)
Congratulations to Ferit Akova for successfully defending his PhD Thesis entitled "A Non-parametric Bayesian Perspective for Machine Learning in Partially-observed Settings".

CAREER Grant Awarded (3/13)
CAREER Grant “CAREER: Self-adjusting Models as a New Direction in Machine Learning” was awarded by NSF.

NIH Grant Awarded (7/12)
Grant with Bartek Rajwa “Automated spectral data transformations and analysis pipeline for high-throughput flow cytometry” was awarded by NIH/NIBIB.