Personal Websites

Maxim Panov

Maxim received his Bachelor and Master degrees from the Moscow Institute of Physics and Technology in 2010 and 2012, respectively. His Bachelor thesis addressed the problem of filling missing data values in the support vector machine classification framework. The Master thesis focused on the Gaussian processes regression and adaptive design of experiments. In 2012, Maxim started his postgraduate studies and switched the direction to the field of Mathematical Statistics. He concentrated on obtaining tight bounds for the Gaussian approximation of posterior distribution (Bernstein-von Mises phenomenon). The research resulted in a series of publications in peer-reviewed journals and formed the core of Maxim’s Candidate of Sciences thesis defended at the Institute for Information Transmission Problems in January 2016.

Starting in 2010, Maxim also worked part time as a research scientist in DATADVANCE Company, which is a resident of Skolkovo Innovation Center. There he participated in developing the library of data analysis methods for engineering applications. This library, pSeven, is now used by a number of companies worldwide, including Airbus, Porsche, Mitsubishi, Toyota, Limagrain and many others.


  • Ivan Nazarov, Maria Burkina, Boris Shirokikh, Gennady Fedonin and Maxim Panov “Sparse-group inductive matrix completion”, submitted to Journal of Computational Mathematics and Mathematical Physics, 2019.
  • Aleksandr Artemenkov and Maxim Panov “NCVis: Noise Contrastive Approach for Scalable Visualization”, 2019.

Machine learning

  • Evgeny Tsymbalov, Sergey Makarychev, Alexander Shapeev and Maxim Panov “Deeper Connections between Neural Networks and Gaussian Processes Regression Speed-up Active Learning”, International Conference Joint on Artificial Intelligence, 2019.
  • Konstantin Slavnov and Maxim Panov “Overlapping Community Detection in Weighted Graphs: Matrix Factorization Approach”, Proceedings of the 11th International Conference on Intelligent Data Processing: Theory and Applications, Springer, 2019 (in press).
  • Stanislav Tsepa and Maxim Panov “Constructing Graph Node Embeddings via Discrimination of Similarity Distributions”, IEEE International Conference on Data Mining WorkshopsICDMW, 1050-1053, 2018.
  • Evgenii Tsymbalov, Maxim Panov and Alexander Shapeev “Dropout-based Active Learning for Regression”, Proceedings of 7th International Conference on Analysis of Images, Social Networks and Texts, 247–258, Lecture Notes in Computer Science, Springer, 2018
  • Panov M., Burnaev E., Yarotsky D., Belyaev M., Kapushev E., Vetrov D., Prikhodko P. GTApprox: Surrogate modeling for industrial design // Advances in Engineering Software. 2016. Vol. 102. P. 29-39.
  • Panov M., Burnaev E., Zaytsev A. Regression on the basis of nonstationary Gaussian processes with Bayesian regularization // Journal of Communications Technology and Electronics. 2016. Vol. 61. No. 6. P. 661-671.
  • Burnaev E., Panov M. Adaptive Design of Experiments Based on Gaussian Processes, in: Lecture Notes in Computer Science Vol. 9047: Statistical Learning and Data Sciences: Third International Symposium, SLDS 2015, Egham, UK, April 20-23, 2015, Proceedings. Switzerland : Springer International Publishing, 2015. P. 116-125.
  • Panov M., Tatarchuk A., Mottl V., Windridge D. A Modified Neutral Point Method for Kernel-Based Fusion of Pattern-Recognition Modalities with Incomplete Data Sets, in: Lecture Notes in Computer Science Vol. 6713: Multiple Classifier Systems: 10th International Workshop, MCS 2011, Naples, Italy, June 15-17, 2011. Proceedings. Springer Berlin Heidelberg, 2011. P. 126-136.



  • S. Fedorov, M.Y. Vasilkov, M. Panov, D. Rupasov, A. Rashkovskiy, N.M. Ushakov, T. Kallio, J. Lee, R. Hempelmann, A.G. Nasibulin “Tailoring electrochemical efficiency of hydrogen evolution by fine design of TiOx/RuOx composite cathode architecture”, International Journal of Hydrogen Energy44(21), 10593–10603, 2019.
  • Dmitry Ermilov, Maxim Panov and Yury Yanovich “Automatic Bitcoin Address Clustering”, Proceedings of the 16th IEEE International Conference on Machine Learning and Applications (ICMLA), Cancun, 2017, pp. 461-466.
  • Nikita Mokrov, Maxim Panov, Boris A. Gutman, Joshua I. Faskowitz, Neda Jahanshad and Paul M. Thompson “Simultaneous Matrix Diagonalization for Structural Brain Networks Classification”, Proceedings of the 6th International Conference on Complex Networks and Their Applications, 2017. Studies in Computational Intelligence, volume 689, pp. 1261 – 1270. Springer.
  • statistical inference, semiparametric inference
  • Nonparametric statistics
  • Algorithms and statistical analysis for random graphs
  • Gaussian processes regression
  • Bayesian methods in machine learning and statistics
  • Moscow Government Award for Young Scientists 2018 (together with Evgeny Burnaev and Alexey Zaytsev)
  • RFBR grant for young scientists 2018-2019

Courses I taught:

  • Introduction to Data Science, Skoltech (Fall 2016-2019)
  • Uncertainty Quantification, Skoltech (Fall 2019)
  • Introduction to Data Science, Moscow Institute of Physics and Technology (Spring 2015-2019)
  • Applied Statistics, Moscow Institute of Physics and Technology (Fall 2011-2018) and Higher School of Economics (Fall 2016)
  • Practical work classes in statistics, Moscow Institute of Physics and Technology (Spring 2012-2013)

Also, I am actively involved in coordination and participation in program creation for Data Science and Information Science and Technology MSc programs and Computational and Data Science and Engineering PhD program at Skoltech (2016 – present)

Service for community

Reviewer for

  • Automation and Remote Control
  • Bernoulli
  • International Journal of Machine Learning and Cybernetics
  • Journal of Nonparametric Statistics
  • Neural Information Processing Systems
  • International Conference on Machine Learning
  • International Conference on Learning Representations