Journal Papers + ML Conferences

  1. Communication-Efficient SGD: From Local SGD to One-Shot Averaging
    A. Spiridonoff, A. Olshevsky, Y. Paschalidis
    Proceedings of NeurIPS 2021.
  2. A Sharp Estimate on the Transient Time of Distributed Stochastic Gradient
    Descent
    S. Pu, A. Olshevsky Y. Paschalidis
    IEEE Transactions on Automatic Control, to appear.
  3. Non-asymptotic Concentration Rates in Cooperative Learning Part I: Variational Non-Bayesian Social Learning
    C. Uribe, A. Olshevsky, A. Nedich
    IEEE Transactions on Control of Network Systems, to appear.
  4. Non-asymptotic Concentration Rates in Cooperative Learning Part II: Inference on Compact Hypothesis Sets
    C. Uribe, A. Olshevsky, A. Nedich
    IEEE Transactions on Control of Network Systems, to appear.
  5. Temporal Difference Learning as Gradient Splitting
    R. Liu, A. Olshevsky
    Proceedings of ICML 2021.
  6. Deterministic and Randomized Actuator Scheduling with Guaranteed Performance Bounds
    M. Siami, A. Jadbabaie, A. Olshevsky
    IEEE Transactions on Automatic Control, 2021.
  7. Adversarial Crowdsourcing Through Robust Rank-1 Matrix Completion
    Q. Ma, A. Olshevsky
    Proceedings of NeurIPS 2020.
  8. Gradient Descent for Sparse Rank-One Matrix Completion for Crowdsourced Aggregation of Sparsely Interacting Workers
    Y. Ma, A. Olshevsky, V. Saligrama, C. Szepesvari
    Journal of Machine Learning Research, 2020.
    Conference version was published as Proceedings of ICML 2018
  9. Robust Asynchronous Stochastic Gradient Push: Asymptotically Optimal and Network Independent Performance for Strongly Convex Functions
    A. Spiridonoff, A. Olshevsky, Y. Paschalidis
    Journal of Machine Learning Research,  2020.
  10. Asymptotic Network Independence in Distributed Optimization for Machine Learning
    S. Pu, A. Olshevsky, Y. Paschlidis
    IEEE Signal Processing Magazine,  2020.
  11. Asymptotic Convergence Rate of Alternating Minimization for Rank One Matrix Completion
    R. Liu, A. Olshevsky
    IEEE Control Systems Letters, 2020

    (All papers below have author names in alphabetical order)

  12. Minimax Rate for Pairwise Comparisons in the BTL Model,
    J. Hendrickx, A. Olshevsky, V. Saligrama
    Proceedings of ICML 2020.
  13. Minimax Rank-1 Matrix Factorization
    J. Hendrickx, A. Olshevsky, V. Saligrama
    Proceedings of AISTATS 2020.
  14. On A Relaxation of Time-Varying Actuator Placement
    A. Olshevsky
    IEEE Control System Letters, 2020
  15. Graph Resistance and Learning from Pairwise Comparisons 
    J. Hendrickx, A. Olshevsky, V. Saligrama
    Proceedings of ICML 2019. 
  16. Graph Theoretic Analysis of Belief System Dynamics Under Logic Constraints
    A. Nedic, A. Olshevsky, C. Uribe
    Scientific Reports, 2019.
  17. On The Inapproximability of the Discrete Witsenhausen Problem
    A. Olshevsky
    IEEE Control System Letters, 2019.
  18. Scaling Laws for Consensus Protocols Subject to Noise
    A. Jadbabaie, A. Olshevsky
    IEEE Transactions on Automatic Control, 2019.
  19. Minimal Reachability is Hard to Approximate
    A. Jadbabaie, A. Olshevsky, G. Pappas, V. Tzoumas
    IEEE Transactions on Automatic Control, 2019.
  20. On (Non)Supermodularity of Average Control Energy
    A. Olshevsky
    IEEE Transactions on Control of Network Systems, 2018.
  21. Network Topology and Communication-Computation Tradeoffs in Distributed Optimization
    A. Nedic, A. Olshevsky, M. Rabbat
    Proceedings of the IEEE, 2018.
  22. Federated Learning of Predictive Models from Federated Electronic Health Records
    T. Brisimi, R. Chen, T. Mela, A. Olshevsky, Y. Paschalidis, W. Shi
    International Journal of Medical Informatics, 2018.
    Special issue on Health Data Science.
  23. Achieving Geometric Convergence for Distributed Optimization over Time-Varying Graphs
    A. Nedic, A. Olshevsky, W. Shi
    SIAM Journal on Optimization, 2017.
  24. Linear Time Average Consensus on Fixed Graphs and Implications for Decentralized Optimization and Multi-Agent Control
    A. Olshevsky
    SIAM Journal on Control and Optimization, 2017.
  25. Fast Convergence Rates for Distributed Non-Bayesian Learning
    A. Nedic, A. Olshevsky, C. Uribe
    IEEE Transactions on Automatic Control, 2017.
  26. Distributed Resource Allocation on Dynamic Networks in Quadratic Time
    T. T. Doan, A. Olshevsky
    Systems & Control Letters, 2017.
  27. Stochastic Gradient-Push for Strongly Convex Functions on Time-Varying Directed Graphs
    A. Nedic, A. Olshevsky
    IEEE Transactions on Automatic Control, 2016.
  28. Convergence Time of Quantized Metropolis Consensus over Time-Varying Networks
    T. Basar, S. R. Etesami, A. Olshevsky
    IEEE Transactions on Automatic Control, 2016.
  29. On Symmetric Continuum Opinion Dynamics
    J. M. Hendrickx, A. Olshevsky
    SIAM Journal on Control and Optimization, 2016.
  30. Eigenvalue Clustering, Control Energy, and Logarithmic Capacity
    A. Olshevsky
    Systems & Control Letters, 2016.
  31. On Primitivity of Sets of Matrices
    V. Blondel, R. Jungers, A. Olshevsky
    Automatica, 2015.
  32. Nonuniform Line Coverage from Noisy Scalar Measurements
    P. Davison, N.E. Leonard, A. Olshevsky, M. Schwemmer
    IEEE Transactions on Automatic Control, 2015.
  33. Distributed Optimization over Time-Varying Directed Graphs
    A. Nedic, A. Olshevsky
    IEEE Transactions on Automatic Control2015.
  34. Cooperative Learning in Multi-Agent Systems From Intermittent Measurements
    N. E. Leonard, A. Olshevsky
    SIAM Journal on Control and Optimization, 2015.
  35. Minimal Controllability Problems
    A. Olshevsky
    IEEE Transactions on Control of Networked Systems, 2014.
  36. How to Decide Consensus? A Combinatorial Necessary and Sufficient Condition and a Proof that Consensus is Decidable but NP-hard
    V. Blondel, A. Olshevsky
    SIAM Journal on Control and Optimization, 2014.
  37. Consensus with Ternary Messages
    A. Olshevsky
    SIAM Journal on Control and Optimization, 2014.
  38. Graph Diameter, Eigenvalues, and Minimum-Time Consensus 
    J. M. Hendrickx, R. M. Jungers, A. Olshevsky, G.  Vankeerberghen
    Automatica, 2014.
  39. Nonuniform Coverage Control on the Line
    N.E. Leonard, A. Olshevsky
    IEEE Transactions on Automatic Control, 2013.
  40. Degree Fluctuations and the Convergence Time of Consensus Algorithms
    A. Olshevsky, J.N. Tsitsiklis
    IEEE Transactions on Automatic Control, 2013.
  41. NP-hardness of Deciding Convexity for Quartic Polynomials and Related Problems
    A.A. Ahmadi, A. Olshevsky, P.A. Parrilo, J.N. Tsitsiklis
    Mathematical Programming, 2013.
  42. Convergence Speed in Distributed Consensus and Averaging
    A. Olshevsky, J. N. Tsitsiklis
    SIAM Review, 2011
    This is a revision of paper originally published in SICON, rewritten
    for a more general audience at the invitation of the SIAM Review.
  43. Distributed Anonymous Discrete Function Computation
    J. M. Hendrickx, A. Olshevsky, J.N. Tsitsklis
    IEEE Transactions on Automatic Control, 2011.
    Special issue on Wireless Sensor and Actuator Networks.
  44. A Lower Bound on Distributed Averaging on the Line Graph
    A. Olshevsky, J.N. Tsitsiklis
    IEEE Transactions on Automatic Control, 2011.
  45. Matrix p-Norms are NP-hard to Approximate if p ≠ 1,2,∞ 
    J. M. Hendrickx, A. Olshevsky
    SIAM Journal on Matrix Analysis and Applications, 2010.
  46. On Distributed Averaging Algorithms and Quantization Effects
    A. Nedic, A. Olshevsky, A. Ozdaglar, J.N. Tsitsiklis
    IEEE Transactions on Automatic Control, 2009.
  47. On the NP-Hardness of Checking Matrix Polytope Stability and Continuous-Time Switching Stability
    L. Gurvits, A. Olshevsky
    IEEE Transactions on Automatic Control, 2009.
  48. Convergence Speed in Distributed Consensus and Averaging
    A. Olshevsky, J.N. Tsitsiklis
    SIAM Journal on Control and Optimization, 2009.
    Special issue on Control and Optimization in Cooperative Networks.
  49. On the Nonexistence of Quadratic Lyapunov Functions for Consensus Algorithms
    A. Olshevsky, J.N. Tsitsiklis
    IEEE Transactions on Automatic Control, 2008.
  50. Improved Approximation Algorithms for the Quality of Service Multicast Tree Problem
    M. Karpinski, I. Mandoiu, A. Olshevsky, A. Zelikovsky
    Algorithmica, 2005.
  51. Kharitonov’s Theorem and Bezoutians
    A. Olshevsky, V. Olshevsky
    Linear Algebra and its Applications, 2005.

Ph.D. Thesis 

Efficient Information Aggregation Strategies for Distributed Control and Signal Processing
Dept. of EECS, MIT, Sep. 2010.

Other Conferences and Book Chapters

(In inverse chronological order, author names always ordered alphabetically).

  1. Leakage Certification Revisited: Bounding Model Errors in Side-Channel Security Evaluations
    O. Bronchain, J. M. Hendrickx, C. Massart, A. Olshevsky, F. Standaert
    Proceedings of Crypto 2019, Santa Barbara, USA, 2019.
  2. Improved Convergence Rate for Distributed Resource Allocation
    A. Nedic, A. Olshevsky, W. Shi
    Proceedings of CDC 2018, the IEEE Conference on Decision and Control, Orlando, USA, 2018. 
  3. Decentralized Consensus Optimization and Resource Allocation
    A. Nedic, A. Olshevsky, W. Shi
    in Large Scale and Distributed Optimization, ed. P. Giselsson, A. Rantzer, Springer Lecture Notes in Mathematics, 2018.
  4. Limitations and Tradeoffs in Minimum Input Selection Problems
    A. Jadbabaie, A. Olshevsky, M. Siami
    Proc. ACC 2018, the American Control Conference, Milwaukee, USA, 2017.
  5. Fully Asynchronous Push-Sum With Growing Intercommunication Intervals
    A. Olshevsky, Y. Paschalidis, A. Spiridonoff
    Proc. of ACC 2018, the American Control Conference, Milwaukee, USA, 2017.
  6. Geometrically Convergent Distributed Optimization with Uncoordinated Step-Sizes
    A. Nedic, A. Olshevsky, W. Shi, C. Uribe
    Proc. of ACC 2017, the American Control Conference, Seattle, USA, 2017. 
  7. Distributed Gaussian Learning over Time-Varying Directed Graphs
    A. Nedic, A. Olshevsky, C. Uribe
    Proc. of Asilomar 2016, the 51st Asilomar Conference on Signals, Systems, and Computers, Monterey, USA, 2016.
  8. Fast Algorithms for Distributed Optimization and Hypothesis Testing: A Tutorial
    A. Olshevsky
    Proc. of CDC 2016, the 55th IEEE Conference on Decision and Control, Las Vegas, USA, 2016.
  9. A Tutorial on Distributed (Non-Bayesian) Learning: Problem, Algorithm, and Results
    A. Nedic, A. Olshevsky, C. Uribe
    Proc. of CDC 2016, the 55th IEEE Conference on Decision and Control, Las Vegas, USA, 2016.
  10. A Geometrically Convergent Method for Distributed Optimization over Time-Varying Graphs
    A. Nedic, A. Olshevsky, W. Shi
    Proc. of CDC 2016, the 55th IEEE Conference on Decision and Control, Las Vegas, USA, 2016.
  11. Distributed Learning with Infinitely Many Hypotheses
    A. Nedic, A. Olshevsky, C. Uribe
    Proc. of CDC 2016, the 55th IEEE Conference on Decision and Control, Las Vegas, USA, 2016.
  12. On Performance of Consensus Protocols Subject to Noise: Role of Hitting Times and Network Structure
    A. Jadbabaie, A. Olshevsky
    Proc. of CDC 2016, the 55th IEEE Conference on Decision and Control, Las Vegas, USA, 2016.
  13. Linearly Convergent Decentralized Consensus Optimization over Directed Networks
    A. Nedic, A. Olshevsky. W. Shi
    Proc. of GlobalSIP 2016, the IEEE Conferencence on Signal and Information Processing, Washington DC, USA, 2016.
  14. Algorithms and Intractability Results for Minimal Controllability Problems
    A. Olshevsky
    Proc. MTNS 2016, the 22nd International Symposium on Mathematics of Networks and Systems, Minneapolis, USA, 2016.
  15. Nonasymptotic Convergence Rates for Cooperative Learning over Time-Varying Directed Graphs
    A. Nedic, A. Olshevsky, C. Uribe
    Proc. ACC 2015, the American Control Conference, Chicago, USA, 2015.
  16. Minimum Input Selection for Structural Controllability
    A. Olshevsky
    Proc. ACC 2015, the American Control Conference, Chicago, USA, 2015.
  17. Linear Time Average Consensus
    A. Olshevsky
    Proc. NecSys 2015, the 5th IFAC Conference on Distributed Estimation and Control in Networked Systems, Philadelphia, USA, 2015.
  18. Fast Convergence of Quantized Consensus Using Metropolis Weights
    T. Basar, S. Etesami, A. Olshevsky
    Proc. CDC 14, the 53rd IEEE Conference on Decision and Control, Los Angeles, USA, 2014.
  19. Focused First-Followers Accelerate Aligning Followers with the Leader in Reaching Network Consensus
    M. Cao, A. Olshevsky, W. Xia
    Proc. of the 19th IFAC World Congress, Cape Town, South Africa, 2014.
  20. Distributed Optimization over Time-Varying Directed Graphs
    A. Nedic, A. Olshevsky
    Proc. CDC 13the 52nd IEEE Conference on Decision and Control, Florence, Italy, 2013.
  21. On Primitivity of Sets of Matrices
    V. Blondel, R. Jungers, A. Olshevsky
    Proc. CDC 13the 52nd IEEE Conference on Decision and Control, Florence, Italy, 2013.
  22. Cooperative Learning in Multi-Agent Systems from Intermittent Measurements
    N. E. Leonard, A. Olshevsky
    Proc. CDC 13the 52nd IEEE Conference on Decision and Control, Florence, Italy, 2013.
  23. Symmetric Continuum Opinion Dynamics: Convergence, but Sometimes Only in Distribution
    J. M. Hendrickx, A. Olshevsky
    Proc. CDC 13the 52nd IEEE Conference on Decision and Control, Florence, Italy, 2013.
  24. Consensus with Ternary Messages
    A. Olshevsky
    Proc. CDC 13the 52nd IEEE Conference on Decision and Control, Florence, Italy, 2013.
  25. Distributed Optimization of Strongly Convex Functions over Time-Varying Graphs
    A. Nedic, A. Olshevsky
    Proc. GlobalSIP 2013, the 1st IEEE Conference on Signal and Information Processing, Austin, USA, 2013.
  26. Combinatorial Bounds and Scaling Laws for Noise Amplification in Networks
    A. Jadbabaie, A. Olshevsky
    Proc. ECC 13, the European Control Conference, Zurich, Switzerland, 2013.
  27. On the Cost of Deciding Consensus
    V. Blondel, A. Olshevsky
    Proc. CDC 12the 51st IEEE Conference on Decision and Control, Maui, HI, 2012.
  28. Nonuniform Coverage Control on the Line
    N.E. Leonard, A. Olshevsky
    Proc. CDC 11the 50th IEEE Conference on Decision and Control, Orlando, FL, 2011.
  29. Degree Fluctuations and the Convergence Times of Consensus Algorithms
    A. Olshevsky, J.N. Tsitsiklis
    Proc. CDC 11the 50th IEEE Conference on Decision and Control, Orlando, FL, 2011.
  30. A Lower Bound on Distributed Averaging on the Line Graph
    A. Olshevsky, J.N. Tsitsiklis
    Proc. CDC 10, the 49th IEEE Conference on Decision and Control, Atlanta, GA, 2010.
  31. Distributed Anonymous Function Computation in Information Fusion and Multiagent Systems
    J. M. Hendrickx, A. Olshevsky, J.N. Tsitsiklis
    Proc. Allerton 09, the Forty-Seventh Allerton Conference on Communication, Control, and Computing, Monticello, Illinois, 2009.
  32. Distributed Subgradient Methods and Quantization Effects
    A. Nedic, A. Olshevsky, A. Ozdaglar, J.N. Tsitsiklis
    Proc. CDC 08, the 47th IEEE Conference on Decision and Control, Cancun, Mexico, 2008.
  33. On Distributed Averaging Algorithms and Quantization Effects
    A. Nedic, A. Olshevsky, A. Ozdaglar, J.N. Tsitsiklis,
    Proc. CDC 08, the 47th IEEE Conference on Decision and Control, Cancun, Mexico, 2008.
  34. Stability Testing of Matrix Polytopes
    L. Gurvits, A. Olshevsky,
    Proc. of ECC 07, the European Control Conference, Kos, Greece, 2007.
  35. Convergence Speed in Distributed Consensus and Averaging
    A. Olshevsky, J.N. Tsitsiklis,
    Proc. CDC 06, the 45th IEEE Conference on Decision and Control, San Diego, USA, 2006.
  36. Convergence in Multiagent Coordination, Consensus, and Flocking
    V.D. Blondel, J.M. Hendrickx, A. Olshevsky, J.N. Tsitsiklis
    Proc. CDC 05, the 44th IEEE Conference on Decision and Control, Seville, Spain, 2005.
  37. Quality of Service in Multimedia Multicast Routing
    I. Mandoiu, A. Olshevsky, A. Zelikovsky
    in Approximation Algorithms and Metaheuristics, T. Gonzales (editor), Chapman and Hall, 2007.
  38. Primal-Dual Algorithms for QoS Multimedia Multicast
    G. Calinescu, C. Fernaneds, I. Mandoiu, A. Olshevsky, K. Yang, A. Zelikovsky
    Proc. GLOBECOM 03, the IEEE Global Communication Conference, San Francisco, USA, 2003.
  39. A Comrade-Matrix-Based Derivation of the Different Version of Fast Cosine and Sine Transforms
    A. Olshevsky, V. Olshevsky, J. Wang
    Proc. SPIE 03, Advanced Signal Processing Algorithms, Architectures, and Implementations XII, 2003.
  40. Improved Approximation Algorithms for the Quality of Service Steiner Tree Problem
    M. Karpinski, I. Mandoiu, A. Olshevsky, A. Zelikovsky
    Proc. WADS 03, the Workshop and Algorithms and Data Structures, Ottawa, Canada, 2003
  41. Network Lifetime and Power Assignment for Ad-Hoc Wireless Networks
    G. Galinescu, S. Kapoor, A. Olshevsky, and A. Zelikovsky
    Proc. ESA 03, the European Sympoisum on Algorithms, Budapest, Hungary, 2003.