CV




I am a PhD candidate in
Computer Science (EconCS group) at Harvard University. My primary advisor is David Parkes (computer science, SEAS) and my secondary advisor is Edo Airoldi (statistics). I received a Master's degree in Statistics from Harvard Univeristy and BS in Electrical Engineering from Sharif University of Technology.

I have been selected as a Siebel Scholar, Class of 2014.

Email: azari -at- fas.harvard.edu
Office: 242 Maxwell-Dworkin, 33 Oxford St, Cambridge MA


| Research | Publications | Presentations | Teaching | Awards | Experience | Services & Leadership | Software
|Good Reads |


Research



My research is in the area of machine learning and statistical modeling with applications in social computation and social networks, where I design scalable algorithms for analyzing large data sets. I have worked on a fast method, called Graphlets, for decomposing large social networks into interpretable components. My current research focuses on designing scalable algorithms for analyzing rank data in applications such as social choice, voting, demand estimation, crowdsourcing and recommender systems.

Publications




Hossein Azari Soufiani, Denis J. Charles, David M. Chickering, David C. Parkes Approximating the Shapley Value via Multi-Issue Decomposition, Accepted in Proceedings of the 13th International Conference on Autonomous Agents and Multiagent Systems (AAMAS 2014). (pdf)



Hossein Azari Soufiani, David C. Parkes, Lirong Xia, Computing Parametric Ranking Models via Rank-Breaking, Accepted for Proceedings of the International Conference on Machine Learning (ICML 2014), Beijing, China. (pdf) (R-Code)

Hossein Azari Soufiani, William Chen , David C. Parkes, Lirong Xia , Generalized Method-of-Moments for Rank Aggregation , Proceedings of the Annual Conference on Neural Information Processing Systems (NIPS 2013), Lake Tahoe, Nevada, USA. (pdf) (R-Code)




Hossein Azari Soufiani, Hansheng Diao , Zhenyu Lai, David C. Parkes, Generalized Random Utility Models with Multiple Types , Proceedings of the Annual Conference on Neural Information Processing Systems (NIPS 2013), Lake Tahoe, Nevada, USA. (pdf) (Supplementry)






Hossein Azari Soufiani, David C. Parkes, Lirong Xia, Preference Elicitation For General Random Utility Models, Proceedings of the 29th Conference on Uncertainty in Artificial Intelligence (UAI 2013), Bellevue, Washington, USA. (pdf) (Poster)






Hossein Azari Soufiani, David C. Parkes, Lirong Xia, Random Utility Theory for Social Choice, Proceedings of the Annual Conference on Neural Information Processing Systems (NIPS 2012), pp. 126-134, Lake Tahoe, Nevada, USA. (pdf) (arXiv version) (Poster) (R-Code)





Sasha Goodman , David Lazer , Edoardo M. Airoldi and Hossein Azari Soufiani, Tracing the Invisible Networks of US Politics, at the 33rd Annual Sunbelt Social Network Conference of the International Network for Social Network Analysis, Hamburg, Germany, May 2013.

Hossein Azari Soufiani, Edoardo M. Airoldi, Graphlets Decomposition of a Weighted Network, Journal of Machine Learning Research, W&CP, 22 (Proceedings of the 15th Internatioanal Conference on Artificial Intelligence and Statistics AISTATS), 54-63. 2012. (pdf) (arXiv version) (Poster) (R-Code) (Recipient of the best student paper award, NESS 2012) .


Hossein Azari Soufiani, M.J. Saberian , M.A. Akhaee, R.N. Mahallati, F. Marvasti, Analysis of communication systems using iterative methods based on Banach's contraction principle, Proceedings of 6th International Conference on Information, Communications and Signal Processing, December 2007 (pdf).



Presentations



Statistical Rank Aggregation through Random Utility Models, Invited talk at 1st annual PhD Symposium Amazon 2013, Seattle WA

Preference Elicitation For General Random Utility Models, Poster presentation at ACM Conference on Electronic Commerce 2013, Philadelphia, PA

General Random Utility Models: Computational Issues, Short talk at 2nd Cambridge Area Economics and Computation Day CAEC 2013, Cambridge, MA

General Random Utility Models, Poster presentation at NEMLí13 and NESSí13, Cambridge and Connecticut, MA

Graphlet decomposition of a weighted network, presnetation at NIPS 2012 Workshop: Algorithmic and Statistical Approaches for Large Social Networks, Lake Tahoe, NV (Slides).

Graphlets decomposition of a weighted network, Poster presented at WIDS@LIDS: Interdisciplinary Workshop on Information and Decision in Social Networks, MIT, October 2012.

Random Utility for Social Choice, Poster presented at CAOSS 2012: Workshop on Computational and Online Social Science, Columbia University, October 2012.

Random Utility for Social Choice, presentation at CS-ECON Seminar series, Duke University, August 2012.

Graphlets decomposition of a weighted network, presentation at NESS 2012, Boston University, April 2012 (Slides: 5min Presentation) .

Quantitatively Evaluating Social Disorder in Union County, SC 1850-1880 (Social Network Analysis Approaches to Nineteenth-Century Political Structures) 36th Annual Meeting of the Social Science History Association, 17-20 November, 2011.

Exploring Three Moments of Crisis in the Criminal Subculture of Union County, South Carolina: 1856, 1859, 1870, Networks and Network Analysis for the Humanities: Reunion Conference An NEH Institute for Advanced Topics in Digital Humanities, October 20 - 22, 2011.

Graphlets: A semiparametric method for analyzing social and information networks with edge weights, SAMSI Complex Networks Transition Workshop, - June 6-7, 2011 and WISE (Workshop on Infusing Statistics and Engineering) June 5-6, 2011.

Statistical Network Modeling for Humanities. Networks and Network Analysis for the Humanities: An NEH Institute for Advanced Topics in Digital Humanities, August 15 - 27, 2010.

Teaching Assistantships



Harvard University, School of Engineering and Applied Sciences

  Computer Science 181. Intelligent Machines: Perception, Learning, and Uncertainty, Spring 2013 (Head TF)
  Computer Science 186. Economics and Computation, Spring 2013 (Head TF)
  Applied Mathematic 101. Statistical Inference for Scientists and Engineers, Fall 2012
  Applied Mathematics 111. Introduction to Scientific Computing, Spring 2012
  Statistics 111. Introduction to Theoretical Statistics, Spring 2012
  Computer Science 281 . Advanced Machine Learning, Fall 2011
  Applied Mathematics 21b. Mathematical Methods in the Sciences, Spring 2010

Sharif University, Electrical Engineering Department

  Measurements & Instrumentation, Spring 2008
  Probability and Statistics, Fall 2007
  Electronics II, Fall 2007
  Electrical Circuits II, Spring 2007
  Electronics I, Fall 2006

Awards



Siebel Scholarship, Class of 2014.

Best student paper award, New England Statistics Symposium (NESS) 2012.

Ranked 2nd out of 190 member class of 2008, Sharif University of Technology, 2008.

Ranked 17th in Iran Nationwide University Entrance Exam, 2004.

Experience



Microsoft Research Redmond, Machine Learning Group: Summer Intern (June 2013 - August 2013)

Harvard University, School of Engineering and Applied Sciences: Research Assistant (Summer 2009 - present)

UCLA, Institute for Pure and Applied Mathematics (IPAM): Teaching Fellow, Networks and Network Analysis for the Humanities (Aug 2010)

AT&T Shannon Laboratory, Statistics Department: Summer Intern (June 2010-July 2010)

Sharif University of Technology, Advanced Communication Research Institute: Research Assistant (2006 - 2008)

Services & Leadership



Supervising a senior undergraduate student in Statistics and a Masters student in Computaitonal Science and Engineering.

Reviewed papers for: IEEE Transactions on Cybernetics, IEEE Transactions on Knowledge and Data Engineering, Statistics and Computing Journal, SODA, JMLR, ICML and AAAI.

Harvard University, Graduate Dormitory Council : President (June 2010-June 2011)

Software



StatRank package for R project.

Good Reads



These are some nice papers:

Economic Choices
The Mathematization of Economic Theory
Independence of Irrelevant Alternative
Uniqueness of the Shapley Value

Expectation Maximization Algorithm
Causal Inference Using Potential Outcomes
Significance of Non-Significance