Generic No Photo Available Image

Mehrdad Mahdavi

Assistant Professor

Affiliation(s):

  • School of Electrical Engineering and Computer Science
  • Computer Science and Engineering

Research Areas:

Data Science and Artificial Intelligence

Interest Areas:

Large-scale optimization, High-dimensional learning, Convex and non-convex Optimization, Distributed optimization, Statistical and computational learning theory, Online (adversarial) learning, Applications of machine learning in social graphs, recommender systems, text analysis, etc

 
 

 

Education

  • Ph D, Michigan State University, 2014
  • Research Assistance Professor, Toyota Technological Institute at Chicago, 2016

Publications

Book, Chapters

  • Rana Forsati and Mehrdad Mahdavi, 2010, Web text mining using harmony search, Springer, Berlin, Heidelberg, pp. 51--64
  • Mehrdad Mahdavi, Rana Forsati and Ali Movaghar, 2009, Bandwidth-delay constrained least cost multicast routing for multimedia communication, Springer Berlin Heidelberg, pp. 737--740
  • Mehrdad Mahdavi, 2009, Solving NP-complete problems by harmony search, Springer Berlin Heidelberg, pp. 53--70

Journal Articles

  • Jialei Wang, Jason D Lee, Mehrdad Mahdavi, Mladen Kolar, Nathan Srebro and others, 2018, "Sketching meets random projection in the dual: A provable recovery algorithm for big and high-dimensional data", Electronic Journal of Statistics, 11, (2), pp. 4896--4944
  • Mehrdad Mahdavi, 2015, "Random Projections for Classification: A Recovery Approach"
  • Tianbao Yang, Mehrdad Mahdavi, Rong Jin and Shenghuo Zhu, 2014, "An efficient primal dual prox method for non-smooth optimization", Machine Learning Journal, pp. 1--38
  • Rana Forsati, Mehrdad Mahdavi, Mehrnoush Shamsfard and Mohamed Sarwat, 2014, "Matrix factorization with explicit trust and distrust side information for improved social recommendation", ACM Transactions on Information Systems (TOIS), 32, (4), pp. 17
  • L Zhang, Mehrdad Mahdavi, R Jin, Tinglu Yang and S Zhu, 2014, "Random Projections for Classification: A Recovery Approach", IEEE Transactions on Information Theory
  • Tianbao Yang, Mehrdad Mahdavi, Rong Jin and Shenghuo Zhu, 2014, "Regret bounded by gradual variation for online convex optimization", Machine Learning, pp. 1--41
  • Rana Forsati, Mehrdad Mahdavi, Mehrnoush Shamsfard and Mohammad Reza Meybodi, 2013, "Efficient stochastic algorithms for document clustering", Information Sciences, 220, pp. 269--291
  • Masri Ayob, Mohammed Hadwan, Mohd Zakree Ahmad Nazri, Zulkifli Ahmad, MA Al-Betar, AT Khader, U Aickelin, KA Dowsland, U Aickelin, EK Burke and others, 2012, "Self-adaptive harmony search algorithm for optimization.", Journal of Applied Sciences, 13, (6), pp. 3--31
  • Mehrdad Mahdavi, Rong Jin and Tianbao Yang, 2012, "Trading regret for efficiency: online convex optimization with long term constraints", Journal of Machine Learning Research (JMLR), 13, pp. 2503--2528
  • Mehrdad Mahdavi and Hassan Abolhassani, 2009, "Harmony k-means algorithm for document clustering", Data Mining and Knowledge Discovery, 18, (3), pp. 370--391
  • Mahamed GH Omran and Mehrdad Mahdavi, 2008, "Global-best harmony search", Applied mathematics and computation, 198, (2), pp. 643--656
  • Mehrdad Mahdavi, M Fesanghary and E Damangir, 2007, "An improved harmony search algorithm for solving optimization problems", Applied Mathematics and Computation, 188, (2), pp. 1567--1579
  • Kamyar Khodamoradi, Mehrdad Mahdavi and Mohammad Ghodsi, 2007, "Upper bounding the price of anarchy in atomic splittable selfish routing"

Conference Proceedings

  • Samet Oymak, Mehrdad Mahdavi and Jiasi Chen, 2019, "Learning Feature Nonlinearities with Non-Convex Regularized Binned Regression", IEEE International Symposium on Information Theory (ISIT)
  • Ofer Meshi, Mehrdad Mahdavi, Adrian Weller and David Sontag, 2016, "Train and Test Tightness of LP Relaxations in Structured Prediction", ICML
  • Mehrdad Mahdavi, Lijun Zhang and Rong Jin, 2015, "Lower and upper bounds on the generalization of stochastic exponentially concave optimization", COLT, pp. 1305--1320
  • Ofer Meshi, Mehrdad Mahdavi and Alex Schwing, 2015, "Smooth and strong: MAP inference with linear convergence", NeurIPS, pp. 298--306
  • Tianbao Yang, Yu-Feng Li, Mehrdad Mahdavi, Rong Jin and Zhi-Hua Zhou, 2013, "Nystrom method vs random fourier features: a theoretical and empirical comparison", NeurIPS, Advances in Neural Information Processing Systems
  • Lijun Zhang, Mehrdad Mahdavi and Rong Jin, 2013, "Linear convergence with condition number independent access of full gradients", NeurIPS, pp. 980--988
  • Mehrdad Mahdavi, Lijun Zhang and Rong Jin, 2013, "Mixed optimization for smooth functions", Advances in Neural Information Processing Systems (NeurIPS), pp. 674--682
  • Mehrdad Mahdavi and Rong Jin, 2013, "Passive learning with target risk", COLT
  • Mehrdad Mahdavi, Tianbao Yang and Rong Jin, 2013, "Stochastic convex optimization with multiple objectives", NeurIPS, pp. 1115--1123
  • Chao-Kai Chiang, Tianbao Yang, Chia-Jung Lee, Mehrdad Mahdavi, Chi-Jen Lu, Rong Jin and Shenghuo Zhu, 2012, "Online optimization with gradual variations.", Conference on Learning Theory (COLT), 23, pp. 6--1
  • Mehrdad Mahdavi, Tianbao Yang and Rong Jin, 2012, "Online decision making under stochastic constraints"
  • Tianbao Yang, Mehrdad Mahdavi, Rong Jin, Jinfeng Yi and Steven CH Hoi, 2012, "Online kernel selection: algorithms and evaluations.", AAAI
  • Lijun Zhang, Mehrdad Mahdavi, Rong Jin and Tianbao Yang, 2012, "Recovering the optimal solution by dual random projection"
  • Jinfeng Yi, Tianbao Yang, Rong Jin, Anil K Jain and Mehrdad Mahdavi, 2012, "Robust ensemble clustering by matrix completion", pp. 1176--1181
  • Mehrdad Mahdavi, Tianbao Yang, Rong Jin, Shenghuo Zhu and Jinfeng Yi, 2012, "Stochastic gradient descent with only one projection", Advances in Neural Information Processing Systems (NeurIPS)
  • Rana Forsati, Mehrdad Mahdavi, Abolfazl Torghy Haghighat and Azadeh Ghariniyat, 2008, "An efficient algorithm for bandwidth-delay constrained least cost multicast routing", pp. 001641--001646
  • Rana Forsati, MohammadReza Meybodi, Mehrdad Mahdavi and AzadehGhari Neiat, 2008, "Hybridization of k-means and harmony search methods for web page clustering", pp. 329--335
  • Rana Forsati, Mehrdad Mahdavi, Mohammadreza Kangavari and Banafsheh Safarkhani, 2008, "Web page clustering using harmony search optimization", pp. 001601--001604
  • Jialei Wang, Jason Lee, Mehrdad Mahdavi, Mladen Kolar and Nathan Srebro, , "A provable recovery algorithm for big and high-dimensional data", AISTATS

Technical Reports

  • Lijun Zhang, Mehrdad Mahdavi and Rong Jin, 2017, "Beating the minimax rate of active learning with prior knowledge", arXiv preprint arXiv:1311.4803
  • Ofer Meshi, Mehrdad Mahdavi and David Sontag, 2015, "On the tightness of lp relaxations for structured prediction", arXiv preprint arXiv:1511.01419
  • Shenghuo Zhu, Rong Jin, Qi Qian, Lijun Zhang, Mehrdad Mahdavi and Tianbao Yang, 2015, "Relational learning using bilinear models and its application in E-commerce"
  • Mehrdad Mahdavi, Lijun Zhang and Rong Jin, 2014, "Binary excess risk for smooth convex surrogates", arXiv preprint arXiv:1402.1792
  • Mehrdad Mahdavi and Rong Jin, 2014, "Excess risk bounds for exponentially concave losses", arXiv preprint arXiv:1401.4566
  • Mehrdad Mahdavi and Rong Jin, 2013, "MixedGrad: an O(1/T) convergence rate algorithm for stochastic smooth optimization", arXiv preprint arXiv:1307.7192
  • Rong Jin, Tianbao Yang and Mehrdad Mahdavi, 2013, "Sparse multiple kernel learning with geometric convergence rate", arXiv preprint arXiv:1302.0315
  • Mehrdad Mahdavi, Lijun Zhang and Rong Jin, 2013, "Supplementary material for mixed optimization for smooth functions"
  • Mehrdad Mahdavi, Tianbao Yang, Rong Jin, Shenghuo Zhu and Jinfeng Yi, 2012, "Supplementary material: stochastic gradient descent with only one projection"
  • Tianbao Yang, Mehrdad Mahdavi, Rong Jin, Lijun Zhang and Yang Zhou, 2012, "Multiple kernel learning from noisy labels by stochastic programming"
  • Mehrdad Mahdavi, Tianbao Yang and Rong Jin, 2012, "An improved bound for the Nystrom method for large eigengap", arXiv preprint arXiv:1209.0001
  • Mehrdad Mahdavi, Tianbao Yang and Rong Jin, 2012, "Efficient constrained regret minimization", arXiv preprint arXiv:1205.2265
  • Mehrdad Mahdavi and Rong Jin, 2012, "High probability bounds for bandit stochastic and adversarial strongly convex optimization"
  • Rong Jin, Tianbao Yang, Mehrdad Mahdavi, Y Li and Z Zhou, 2012, "Improved bounds for the Nystrom method with application to kernel classification"

Other

  • Mehrdad Mahdavi, 2014, "Exploiting Smoothness in Statistical Learning, Sequential Prediction, and Stochastic Optimization"

Research Projects

Honors and Awards

  • Mark Fulk Best Student Paper Award, 2012 - 2012
  • Top Cited Paper Award, Elsevier, 2010 - 2010

Service

Service to Penn State:

Service to External Organizations:

 


 

About

The School of Electrical Engineering and Computer Science was created in the spring of 2015 to allow greater access to courses offered by both departments for undergraduate and graduate students in exciting collaborative research in fields.

We offer B.S. degrees in electrical engineering, computer science, computer engineering and data science and graduate degrees (master's degrees and Ph.D.'s) in electrical engineering and computer science and engineering. EECS focuses on the convergence of technologies and disciplines to meet today’s industrial demands.

School of Electrical Engineering and Computer Science

The Pennsylvania State University

207 Electrical Engineering West

University Park, PA 16802

814-863-6740

Department of Computer Science and Engineering

814-865-9505

Department of Electrical Engineering

814-865-7667