Biography

I’m a staff research scientist in the Theoretical Division at the Los Alamos National Laboratory (LANL), New Mexico, where I am part of the Advanced Network Science Initiative (ANSI) as well as the Condensed Matter and Complex Systems Group (T-4). My background is in statistical physics and information theory.

My current work focuses on the design of machine learning techniques for learning probabilistic networks and on the development of new methods to control and optimize energy networks under uncertainty.

Machine Learning

• Efficient Learning of Discrete Graphical Models
  M. Vuffray, S. Misra, A.Y. Lokhov
  arXiv, 2019, [arXiv]

• Information Theoretic Optimal Learning of Gaussian Graphical Models
  S. Misra, M. Vuffray, A.Y. Lokhov
  arXiv, 2019, [arXiv]

• Optimal Structure and Parameter Learning of Ising Models
  A.Y. Lokhov, M. Vuffray, S. Misra, M. Chertkov
  Science Advances, 2018, [online], [arXiv]

• Interaction Screening: Efficient and Sample-Optimal Learning of Ising Models
  M. Vuffray, S. Misra, A.Y. Lokhov, M. Chertkov
  Advances in Neural Information Processing Systems (NeurIPS), 2016, [online], [arXiv]

Energy Networks

• Uncovering Power Transmission Dynamic Model from Incomplete PMU Observations
  A.Y. Lokhov, D. Deka, M. Vuffray, M. Chertkov
  IEEE Conference on Decision and Control (CDC), 2018, [online]

• Online Learning of Power Transmission Dynamics
  A.Y. Lokhov, M. Vuffray, D. Shemetov, D. Deka, M. Chertkov
  Power Systems Computation Conference (PSCC), 2018, [online], [arXiv]

• Graphical Models for Optimal Power Flow
  K. Dvijotham, M. Chertkov, P. Van Hentenryck, M. Vuffray, S. Misra
  Constraints, 2017, [online], [arXiv]

• Monotonicity of Dissipative Flow Networks Renders Robust Maximum Profit Problem Tractable: General Analysis and Application to Natural Gas Flows
  M. Vuffray, S. Misra, M. Chertkov
  IEEE Conference on Decision and Control (CDC), 2015, [online], [arXiv]

Information Theory

• The Bethe Free Energy Allows to Compute the Conditional Entropy of Graphical Code Instances: A Proof From the Polymer Expansion
  N. Macris, M. Vuffray
  IEEE Transactions on Information Theory, 2016, [online], [arXiv]

• Approaching the Rate-Distortion Limit with Spatial Coupling, Belief Propagation, and Decimation
  V. Aref, N. Macris, M. Vuffray
  IEEE Transactions on Information Theory, 2015, [online], [arXiv]