Professor Shahadat Hossain

Biography

After receiving a B.Tech. in Textiles from the University of Dhaka, Bangladesh, I availed an opportunity to study computational optimization in the Department of Informatics, University of Bergen, Norway under Professor Trond Steihaug. I was awarded a University of Bergen Fellowship (Universitetsstipendiat) for my doctoral studies during which (1996) I was a visiting graduate student of Cornell CS department. Before coming to Lethbridge I was on the faculty of CS department of the University of Northern British Columbia. I spent a part of my sabbatical leave (in 2006) at Humboldt University in Berlin, Germany. My academic ancestry can be traced back, starting at Forsythe Tree, all the way to Gottfried Wilhelm Leibniz  and beyond at The Mathematics Genealogy Project.

Contact Information

 

Teaching (Fall 2017)

Research.

I am interested in numerical optimization (broadly interpreted) and its applications. Scientific problems that are large-scale and display “exploitable” special properties such as sparsity and structural information such as symmetry are particularly intriguing. I find graph theory to be an important tool for exploiting the combinatorial structure inherent in many sub-problems in nonlinear and linear optimization. My current research is concerned with the design of efficient algorithms for sparse matrix problems: efficient determination of Jacobian and Hessian matrices, linear algebra of sparse matrices with special properties, and high-performance computing. 

List of publications can be found in Google Scholar and  Research Gate.

Selected Recent Publications (student co-authors in boldface)

Determination of Large Sparse Derivative Matrices: Structural Orthogonality and Structural Degeneracy (5 pages), Shahadat Hossain and Ashraful Suny, 15th Cologne-Twente workshop on graphs and combinatorial optimization, June 6-8, 2017, Cologne, Germany.


Pattern Graph for Sparse Hessian Matrix Determination,  Shahadat Hossain and Nasrin Hakim Mithila, The 7th International Conference on Algorithmic Differentiation 2016, pp. 78 - 81, Christ Church Oxford, UK.

Efficient Detection of Hessian Matrix Sparsity Pattern, Richard Carter, Shahadat Hossain and Marzia Sultana, ACM Communications in Computer Algebra Volume 50 Issue 4, December 2016, pp. 151-154. 

DSMDE: A Data Exchange format for Design Structure Models,  Ansari, Lazima; Hossain, Shahadat;  Khan, Ahamad Imtiaz. Sustainability in Modern Project Management: Proceedings of the 18th International DSM Conference, pp. 111-121, 2016. ISBN 978-85-63710-01-7, DOI 10.19255/JMPM-DSM2016

DSJM: A Software Toolkit for Direct Determination of Sparse Jacobian Matrices (inbook), Hasan, Mahmudul and Hossain, Shahadat and Khan, Ahamad Imtiaz and Mithila, Nasrin Hakim and Suny, Ashraful Huq, in Mathematical Software –ICMS 2016 : Greuel, Gert-Martin and Koch, Thorsten and Paule, Peter and Sommese, Andrew (Editors), pp 275--283 Springer International Publishing, 2016.

Determining Sparse Jacobian Matrices Using Two-Sided Compression: An Algorithm and Lower Bounds (inbook), Gaur, Daya R. and Hossain, Shahadat and Saha, Anik, in Mathematical and Computational Approaches in Advancing Modern Science and Engineering, Bélair, Jacques and Frigaard, A. Ian and Kunze, Herb and Makarov, Roman and Melnik, Roderick and Spiteri, J. Raymond (Editors),  pp. 425--434,  Springer International Publishing, 2016.

Selim Akl, Robert Benkoczi, Daya Ram Gaur, Hossam Hassanein, Shahadat Hossain and  Mark Thom, On a class of covering problems with variable capacities in wireless networks, Theoretical Computer Science, Available online 4 November 2014, ISSN 0304-3975.

Shahadat Hossain and  Trond Steihaug, Graph models and their efficient implementation for sparse Jacobian matrix determination, Discrete Applied Mathematics, Volume 161, Issue 12, August 2013, Pages 1747-1754


Shahadat Hossain, Trond Steihaug 

Optimal direct determination of sparse Jacobian matrices    Optimization Methods and Software   Vol. 28, Issue 6, 2013.


Selected Talks (by invitation).

Computing sparse derivatives and consecutive zeros problem, 55th CORS Annual Conference, Vancouver, Canada,  May 27-29, 2013.

The CPR Method and Beyond, CMS/CSHPM Summer Meeting 2009, Memorial University of Newfoundland, St. John's,Newfoundland, June 6 - 8, 2009.

Pattern Graphs for Sparse Matrices, SIAM Conference on Optimization, Boston, USA , May 10 -- 13, 2008.

On Efficient Data Structures for Sparse Matrix Storage, Computing by the Numbers: Algorithms, Precision, and Complexity, Matheon Workshop 2006 in honor of the 60th birthday of Richard Brent, Weierstrass Institute for Applied Analysis and Stochastics, July 20 - 21, 2006, Berlin.

Software.

DSJMToolkit Determine Sparse Jacobian Matrices Toolkit.

CsegGraph : Column Segment Graph Generator (Journal of Computer Mathematics Taylor and Francis).

(With Minhaz Zibran) M-Sched: A University Course Timetabler, The 7th International Conference on the Practice and Theory of Automated Timetabling, 2008.

Graduate Students.

Undergraduate Research Assistants.

Stephanie McIntyre (2017), MBen Burnett, Jesse Dunn, Meru Brunn, Zhenshuan Zhang, Matt Voronoi, VivekTrivedi

 Research Funding

My research is mainly supported by NSERC Discovery Grant (DG) Individual and NSERC Collaborative Research and Development (CRD) Grant held jointly with Colleagues in the Department of Computing and Software at McMaster University and Sponsored by IBM (Canada).