Publications

[D1] A. Mokhtari. Efficient Methods for Large-Scale Empirical Risk Minimization, Ph.D. Dissertation, University of Pennsylvania, 2017. [pdf]
       [Joseph and Rosaline Wolf Award for Best Doctoral Dissertation]

Preprints

[P7] (α-β order) A. Mokhtari, A. Ozdaglar, and S. Pattathil. Proximal Point Approximations Achieving a Convergence Rate of  O(1/k) for Smooth Convex-Concave Saddle Point Problems: Optimistic Gradient and Extra-gradient Methods [pdf]

[P6] (α-β order) H. Hassani, A. Karbasi, A. Mokhtari, and Z. Shen. Stochastic Conditional Gradient++ [pdf]

[P5] (α-β order) A. Mokhtari, A. Ozdaglar, and S. Pattathil. A Unified Analysis of Extra-gradient and Optimistic Gradient Methods for Saddle Point Problems: Proximal Point Approach [pdf]

[P4] M. Zhang, L. Chen, A. Mokhtari, H. Hassani, and A. Karbasi. Quantized Frank-Wolfe: Communication- Efficient Distributed Optimization [pdf]

[P3] A. Reisizadeh, A. Mokhtari, H. Hassani, and R. Pedarsani. An Exact Quantized Decentralized Gradient Descent Algorithm [pdf]

[P2] A. Mokhtari, H. Hassani, A. Karbasi. Stochastic Conditional Gradient Methods: From Convex Minimization to Submodular Maximization [pdf]

[P1] M. Eisen, A. Mokhtari, and A. Ribeiro. A Primal-Dual Quasi-Newton Method for Exact Consensus Optimization [pdf]

Journal Papers

[J13] S. Paternain, A. Mokhtari, and A. Ribeiro. A Newton-based Method for Nonconvex Optimization with Fast Evasion of Saddle Points, SIAM Journal on Optimization (SIOPT), 2019. [pdf] 

[J12] A. Mokhtari, M. Eisen, and A. Ribeiro. IQN: An Incremental Quasi-Newton Method with Local Superlinear Convergence Rate, SIAM Journal on Optimization (SIOPT), 2018. [pdf]

[J11] A. Mokhtari, M. Gürbüzbalaban, and A. Ribeiro. Surpassing Gradient Descent Provably: A Cyclic Incremental Method with Linear Convergence Rate, SIAM Journal on Optimization (SIOPT), 2018. [pdf]

[J10] A. Simonetto, A. Koppel, A. Mokhtari, G. Leus, and A. Ribeiro. Decentralized Prediction-Correction Methods for Networked Time-Varying Convex Optimization,IEEE Transactions on Automatic Control (TAC), 2017. [pdf]

[J9] T. Chen, A. Mokhtari, X. Wang, A. Ribeiro, and G. B. Giannakis.Stochastic Averaging for Constrained Optimization with Application to Online Resource Allocation,IEEE Transactions on Signal Processing (TSP), 2017. [pdf]

[J8] M. Eisen, A. Mokhtari, and A. Ribeiro. Decentralized Quasi-Newton Methods, IEEE Transactions on Signal Processing (TSP), 2017. [pdf]

[J7] A. Mokhtari, Q. Ling, and A. Ribeiro. Network Newton Distributed Optimization Methods, IEEE Transactions on Signal Processing (TSP), 2017.[pdf]

[J6] A. Mokhtari and A. Ribeiro. DSA: Decentralized Double Stochastic Averaging Gradient Algorithm, Journal of Machine Learning Research (JMLR), 2016. [pdf]

[J5] A. Mokhtari, W. Shi, Q. Ling, and A. Ribeiro. A Decentralized Second-Order Method with Exact Linear Convergence Rate for Consensus Optimization, IEEE Transactions on Signal and Information Processing over Networks (TSIPN), 2016.[pdf]

[J4] A. Mokhtari, W. Shi, Q. Ling, and A. Ribeiro. DQM: Decentralized Quadratically Approximated Alternating Direction Method of Multipliers, IEEE Transactions on Signal Processing (TSP), 2016.[pdf]

[J3] A. Simonetto, A. Mokhtari, A. Koppel, G. Leus, and A. Ribeiro. A Class of Prediction-Correction Methods for Time-Varying Convex Optimization, IEEE Transactions on Signal Processing (TSP), 2016.[pdf]

[J2] A. Mokhtari and A. Ribeiro. Global Convergence of Online Limited Memory BFGS, Journal of Machine Learning Research (JMLR), 2015. [pdf]

[J1] A. Mokhtari and A. Ribeiro. RES: Regularized Stochastic BFGS Algorithm, IEEE Transactions on Signal Processing (TSP), 2014. [pdf]

Machine Learning Conference Papers

[ML-C9] A. Mokhtari, A. Ozdaglar, and A. Jadbabaie. Efficient Nonconvex Empirical Risk Minimization via Adaptive Sample Size Methods,  Int. Conference on Artificial Intelligence and Statistics (AISTATS), 2019. [pdf]

[ML-C8] A. Mokhtari, A. Ozdaglar, and A. Jadbabaie. Escaping Saddle Points in Constrained Optimization, Advances in Neural Information Processing Systems (NIPS), 2018. [pdf] (Spotlight: Top 4% of the submitted papers)

[ML-C7] J. Zhang, A. Mokhtari, S. Sra, and A. Jadbabaie. Direct Runge-Kutta Discretization Achieves Acceleration, Advances in Neural Information Processing Systems (NIPS), 2018. [pdf] (Spotlight: Top 4% of the submitted papers)

[ML-C6] A. Mokhtari, H. Hassani, and A. Karbasi. Decentralized Submodular Maximization: Bridging Discrete and Continuous Settings, Int. Conference on Machine Learning (ICML), 2018. [pdf] [Supplementary material] (Long talk)

[ML-C5] Z. Shen, A. Mokhtari, H. Qian, P. Zhao, and T. Zhou. Towards More Efficient Stochastic Decentralized Learning: Faster Convergence and Sparse Communication, Int. Conference on Machine Learning (ICML), 2018. [pdf] [Supplementary material]  

[ML-C4] A. Mokhtari, H. Hassani, and A. Karbasi. Conditional Gradient Method for Stochastic Submodular Maximization: Closing the Gap, Int. Conference on Artificial Intelligence and Statistics (AISTATS), 2018. [pdf] [Supplementary material]

[ML-C3] M. Eisen, A. Mokhtari, and A. Ribeiro. Large Scale Empirical Risk Minimization via Truncated Adaptive Newton Method, Int. Conference on Artificial Intelligence and Statistics (AISTATS), 2018. [pdf] [Supplementary material] 

[ML-C2] A. Mokhtari and A. Ribeiro. First-Order Adaptive Sample Size Methods to Reduce Complexity of Empirical Risk Minimization, Advances in Neural Information Processing Systems (NIPS), 2017. [pdf] [Supplementary material]

[ML-C1] A. Mokhtari, H. Daneshmand, A. Lucchi, T. Hofmann, and A. Ribeiro. Adaptive Newton Method for Empirical Risk Minimization to Statistical Accuracy, Advances in Neural Information Processing Systems (NIPS), 2016. [pdf] [Supplementary Material]

[SP-C21] A. Koppel, A. Mokhtari, and A. Ribeiro. Parallel Stochastic Successive Convex Approximation Method for Large-Scale Dictionary Learning, Int. Conf. Acoustics Speech Signal Processing (ICASSP), 2018.

[SP-C20] M. Eisen, A. Mokhtari, and A. Ribeiro. A Primal-Dual Quasi-Newton Method for Consensus Optimization, Asilomar Conference on Signals, Systems, and Computers (Asilomar), 2017. [pdf]

[SP-C19] A. Mokhtari, M. Eisen, and A. Ribeiro. An Incremental Quasi-Newton Method with a Local Superlinear Convergence Rate, Int. Conf. Acoustics Speech Signal Processing (ICASSP), 2017. [pdf]

[SP-C18] A. Mokhtari, M. Gürbüzbalaban, and A. Ribeiro. A Double Incremental Aggregated Gradient Method with Linear Convergence Rate for Large-Scale Optimization, Int. Conf. Acoustics Speech Signal Processing (ICASSP), 2017. [pdf]

[SP-C17] A. Mokhtari, A. Koppel, G. Scutari, and A. Ribeiro. Large-Scale NonConvex Stochastic Optimization by Doubly Stochastic Successive Convex Approximation, Int. Conf. Acoustics Speech Signal Processing (ICASSP), 2017. [pdf]

[SP-C16] A. Mokhtari and A. Ingber. A Diagonal-Augmented Quasi-Newton Method with Application to Factorization Machines, Int. Conf. Acoustics Speech Signal Processing (ICASSP), 2017. [pdf]

[SP-C15] T. Chen, A. Mokhtari, X. Wang, A. Ribeiro, and G. B. Giannakis. A Data-driven Approach to Stochastic Network Optimization, IEEE Global Conference on Signal and Information Processing (GlobalSIP), 2016. [pdf]

[SP-C14] H. Zhang, W. Shi, A. Mokhtari, A. Ribeiro, and Q. Ling. Decentralized Constrained Consensus Optimization with Primal-Dual Splitting Projection, IEEE Global Conference on Signal and Information Processing (GlobalSIP), 2016.  [pdf]

[SP-C13] M. Eisen, A. Mokhtari, and A. Ribeiro. An Asynchronous Quasi-Newton Method for Consensus Optimization, IEEE Global Conference on Signal and Information Processing (GlobalSIP), 2016. [pdf]

[SP-C12] A. Mokhtari, W. Shi, and Qing Ling. ESOM: Exact Second-Order Method for Consensus Optimization, Asilomar Conference on Signals, Systems, and Computers (Asilomar), 2016. [pdf]

[SP-C11] A. Koppel, A. Mokhtari, and A. Ribeiro. Doubly Stochastic Algorithms for Large-Scale Optimization, Asilomar Conference on Signals, Systems, and Computers (Asilomar), 2016. [pdf]

[SP-C10] A. Simonetto, A. Mokhtari, A. Koppel, G. Leus, and A. Ribeiro. A Decentralized Prediction-Correction Method for Networked Time-Varying Convex Optimization, IEEE Int. Workshop on Computational Advances in Multi-Sensor Adaptive Processing (CAMSAP), 2015. [pdf]

[SP-C9] A. Mokhtari, W. Shi, Q. Ling, and A. Ribeiro. Decentralized Quadratically Approximated Alternating Direction Method of Multipliers, IEEE Global Conference on Signal and Information Processing (GlobalSIP), 2015.[pdf]

[SP-C8] A. Koppel, A. Simonetto, A. Mokhtari, G. Leus, and A. Ribeiro. Target Tracking with Dynamic Convex Optimization, IEEE Global Conference on Signal and Information Processing (GlobalSIP), 2015. [pdf]

[SP-C7] A. Mokhtari and A. Ribeiro. Decentralized Double Stochastic Averaging Gradient, Asilomar Conference on Signals, Systems, and Computers (Asilomar), 2015. [pdf]

[SP-C6] A. Simonetto, A. Koppel, A. Mokhtari, G. Leus, and A. Ribeiro. Prediction-Correction Methods for Time-Varying Convex Optimization, Asilomar Conference on Signals, Systems, and Computers (Asilomar), 2015. [pdf]

[SP-C5] A. Mokhtari, Q. Ling, and A. Ribeiro. An Approximate Newton Method for Distributed Optimization, Int. Conf. Acoustics Speech Signal Processing (ICASSP), 2015. [pdf]

[SP-C4] A. Mokhtari, Q. Ling, and A. Ribeiro. Network Newton, Asilomar Conference on Signals, Systems, and Computers (Asilomar), 2014. [pdf]