Publications

• Tracing Sedimentary Origins in Multivariate Geochronology via Constrained Tensor Factorization
  N. Graham, N. Richardson, M. P. Friedlander, J. Saylor
  Preprint, May 2024; revised October, 2024

• Cardinality-constrained structured data-fitting problems
  Z. Fan, H. Fang, M. P. Friedlander
  Open J. Math. Optim., 2023

• From perspective maps to epigraphical projection
  M. P. Friedlander, A. Goodwin, T. Hoheisel
  Mathematics of Operations Research (48)3: 1213-1809, 2022, 2022
• Knowledge-injected federated learning
  Z. Fan, H. Fang, Z. Zhou, J. Pei, M. P. Friedlander, J. Hu, C. Li, Y. Zhang
  arXiv:2208.07530, 2022
• Quantum algorithms for structured prediction
  B. Sephehry, E. Iranmanesh, M. P. Friedlander, P. Ronagh
  Quantum Machine Intelligence, 2022
• Polar deconvolution of mixed signals
  Z. Fan, H. Jeong, B. Joshi, M. P. Friedlander
  IEEE Trans. Signal Processing, 2022
• Online mirror descent and dual averaging: keeping pace in the dynamic case
  H. Fang, N J. A. Harvey, V. S. Portella, M. P. Friedlander
  J. Machine Learning Research, 2022
• A dual approach for federated learning
  Z. Fan, H. Fang, M. P. Friedlander
  arXiv 2201.11183, 2022
• Fair and efficient contribution valuation for vertical federated learning
  Z. Fan, H. Fang, Z. Zhou, J. Pei, M. P. Friedlander, Y. Zhang
  arXiv:2201.02658, 2022
• Improving fairness for data valuation in horizontal federated learning
  Z. Fan, H. Fang, Z. Zhou, J. Pei, M. P. Friedlander, C. Liu, Y. Zhang
  Intern Conf Data Engineering (ICDE), 2022

• NBIHT: An efficient algorithm for 1-bit compressed sensing with optimal error decay rate
  M. P. Friedlander, H. Jeong, Y. Plan, O. Yιlmaz
  IEEE Trans Info Theory, 2021
• Fast convergence of the stochastic subgradient method under interpolation
  H. Fang, Z. Fan, M. P. Friedlander
  Intern. Conf. Learning Representations (ICLR), 2021

• Atomic decomposition via polar alignment: the geometry of structured optimization
  Z. Fan, H. Jeong, Y. Sun, M. P. Friedlander
  Foundations and Trends in Optimization, 3(4):280–366, 2020
• Implementing a smooth exact penalty function for equality-constrained nonlinear optimization
  R. Estrin, M. P. Friedlander, D. Orban, M. A. Saunders
  SIAM J. Sci. Comput., 42(3), A1809–A1835, 2020
• Implementing a smooth exact penalty function for general constrained nonlinear optimization
  R. Estrin, M. P. Friedlander, D. Orban, M. A. Saunders
  SIAM J. Sci. Comput., 42(3), A1836–A1859, 2020
• A perturbation view of level-set methods for convex optimization
  R. Estrin, M. P. Friedlander
  Optimization Letters, 2020
• Online mirror descent and dual averaging: keeping pace in the dynamic case
  H. Fang, N J. A. Harvey, V. S. Portella, M. P. Friedlander
  Intern. Conf. Machine Learning (ICML), 2020
• Greed meets sparsity: understanding and improving greedy coordinate descent for sparse optimization
  H. Fang, Z. Fan, Y. Sun, M. P. Friedlander
  Intern. Conf. Artificial Intelligence and Statistics (AISTATS), 2020

• Bundle methods for dual atomic pursuit
  Z. Fan, Y. Sun, M. P. Friedlander
  Asilomar Conf. Signals, Systems, Computers (ACSSC), 2019
• One-shot atomic detection
  Y. Sun, M. P. Friedlander
  IEEE Intern. Workshop Comput. Adv. Multi-Sensor Adaptive Proc. (CAMSAP), 2019
• Fast training for large-scale one-versus-all linear classifiers using tree-structured initialization
  H. Fang, M. Cheng, C.-J. Hsieh, M. P. Friedlander
  SIAM Intern. Conf. Data Mining (SDM), 2019
• Polar convolution
  M. P. Friedlander, I. Macêdo, T.K. Pong
  SIAM J. Optimization, 29(2):1366–1391, 2019
• Smooth structured prediction using quantum and classical gibbs samplers
  B. Sepehry, E. Iranmanesh, M. P. Friedlander, P. Ronagh
  Adiabatic Quantum Computing Conference (AQC), arXiv:1809.04091, 2019

• Foundations of gauge and perspective duality
  A. Y. Aravkin, J. V. Burke, D. Drusvyatskiy, M. P. Friedlander, K. MacPhee
  SIAM J. Optimization, 28(3):2406–2434, 2018
• Level-set methods for convex optimization
  A. Y. Aravkin, J. V. Burke, D. Drusvyatskiy, M. P. Friedlander, S. Roy
  Mathematical Programming, 174(1-2):359–390, 2018

• Efficient evaluation of scaled proximal operators
  M. P. Friedlander, G. Goh
  Electronic Trans. Numerical Analysis, 46:1–22, 2017

• Satisfying real-world goals with dataset constraints
  G. Goh, A. Cotter, M. Gupta, M. P. Friedlander
  Advances in Neural Information Processing Systems 29 (NIPS), 2016
• Social resistance
  M. P. Friedlander, N. Krislock, T. K. Pong
  IEEE Comput. Sci. Eng., 8(2):98-103; reprinted in Computing Edge, 2016
• Low-rank spectral optimization via gauge duality
  M. P. Friedlander, I. Macêdo
  SIAM J. Scientific Computing, 38(3):A1616–A1638, 2016

• Coordinate descent converges faster with the Gauss-Southwell rule than random selection
  J. Nutini, M. Schmidt, I. H. Laradji, M. P. Friedlander, H. Koepke
  Intern. Conf. Machine Learning (ICML), 2015

• Gauge optimization and duality
  M. P. Friedlander, I. Macêdo, T. K. Pong
  SIAM J. Optimization, 24(4):1999–2022, 2014

• Variational properties of value functions
  A. Aravkin, J. V. Burke, M. P. Friedlander
  SIAM J. Optimization, 23(3):1689–1717, 2013
• Fast dual variational inference for non-conjugate latent gaussian models
  M. E. Khan, A. Y. Aravkin, M. P. Friedlander, M. Seeger
  Intern. Conf. Machine Learning (ICML), 2013
• Tail bounds for stochastic approximation
  M. P. Friedlander, G. Goh
  arXiv:1304.5586, 2013

• Hybrid deterministic-stochastic methods for data fitting
  M. P. Friedlander, M. Schmidt
  SIAM J. Scientific Computing, 34(3):A1380–A1405, 2012
• A primal-dual regularized interior-point method for convex quadratic programs
  M. P. Friedlander, D. Orban
  Mathematical Programming Computation, 4(1):71–107, 2012
• Recovering compressively sampled signals using partial support information
  M. P. Friedlander, H. Mansour, R. Saab, Ö. Yılmaz
  IEEE Trans. Info. Theory, 58(2):1122–1134, 2012
• Fighting the curse of dimensionality: compressive sensing in exploration seismology
  F. J. Herrmann, M. P. Friedlander, Ö. Yılmaz
  IEEE Signal Processing Magazine, 29(3):88–100, 2012
• Robust inversion via semistochastic dimensionality reduction
  A. Aravkin, M. P. Friedlander, T. van Leeuwen
  Intern. Conf. Acoustics, Speech, and Signal Processing (ICASSP), 2012
• Robust inversion, dimensionality reduction, and randomized sampling
  A. Aravkin, M. P. Friedlander, F. Herrmann, T. van Leeuwen
  Mathematical Programming, 134(1):101–125, 2012, 2012

• Sparse optimization with least-squares constraints
  E. van den Berg, M. P. Friedlander
  SIAM J. Optimization, 21(4):1201–1229, 2011

• Theoretical and empirical results for recovery from multiple measurements
  E. van den Berg, M. P. Friedlander
  IEEE Trans. Info. Theory, 56(5):2516–2527, 2010

• Optimizing costly functions with simple constraints: a limited-memory projected quasi-Newton algorithm
  M. Schmidt, E. van den Berg, M. P. Friedlander, K. Murphy
  Intern. Conf. Artificial Intelligence and Statistics (AISTATS), 2009
• Sparco: a testing framework for sparse reconstruction
  E. van den Berg, M. P. Friedlander, G. Hennenfent, F. Herrmann, R. Saab, Ö. Yılmaz
  ACM Trans. Math. Software, 35(4):1–16, 2009

• Probing the Pareto frontier for basis pursuit solutions
  E. van den Berg, M. P. Friedlander
  SIAM J. Scientific Computing, 31(2):890–912, 2008
• Computing nonnegative tensor factorizations
  M. P. Friedlander, K. Hatz
  Optimization Methods and Software, 23(4):631–647, 2008
• New insights into one-norm solvers from the Pareto curve
  G. Hennenfent, E. van den Berg, M. P. Friedlander, F. Herrmann
  Geophysics, 73(4):A23–A26, 2008
• Group sparsity via linear-time projection
  E. van den Berg, M. Schmidt, M. P. Friedlander, K. Murphy
  Tech. Rep. TR-2008-09, Dept of Computer Science, UBC, 2008
• Global and finite termination of a two-phase augmented Lagrangian filter method for general quadratic programs
  M. P. Friedlander, S. Leyffer
  SIAM J. Scientific Computing, 30(4):1706–1726, 2008

• Discussion: The Dantzig selector: Statistical estimation when p is much larger than n
  M. P. Friedlander, M. A. Saunders
  Annals of Statistics, 35(6):2385–2391, 2007
• Exact regularization of convex programs
  M. P. Friedlander, P. Tseng
  SIAM J. Optimization, 18(4):1326–1350, 2007
• A filter active-set trust-region method
  M. P. Friedlander, N. I. M. Gould, S. Leyffer, T. S. Munson
  Preprint ANL/MCS-P1456-0907, Argonne National Laboratory, 2007
• In pursuit of a root
  E. van den Berg, M. P. Friedlander
  Tech. Rep. TR-2007-19, Dept. of Computer Science, UBC, 2007
• Diffuse optical fluorescence tomography using time-resolved data acquired in transmission
  F. Leblond, S. Fortier, M. P. Friedlander
  Multimodal Biomedical Imaging II, vol. 6431. Proc. Intern. Society Optimal Imaging, 2007

• On minimizing distortion and relative entropy
  M. P. Friedlander, M. R. Gupta
  IEEE Trans. Info. Theory, 52(1):238–245, 2006

• A two-sided relaxation scheme for mathematical programs with equilibrium constraints
  V. Demiguel, M. P. Friedlander, F. J. Nogales, S. Scholtes
  SIAM J. Optimization, 16(1):587–609, 2005
• A globally convergent linearly constrained Lagrangian method for nonlinear optimization
  M. P. Friedlander, M. A. Saunders
  SIAM J. Optimization 15(3):863–897, 2005

• Maximum entropy classification applied to speech
  M. R. Gupta, M. P. Friedlander, R. M. Gray
  Asilomar Conf. Signals, Systems, Computers (ACSSC), vol. 2, 1480–1483, 2000

Selected Talks

• Polar deconvolution of mixed signals, One-World Optimization Seminar. October, 26, 2020
• Polar Duality in three liftings, McGill University, Montreal. November 28, 2017
• Algorithms for sparse optimization, Google Research, Mountain View. May 19, 2015
• Robust inversion and randomized sampling, International Symposium on Mathematical Programming (ISMP), Berlin. August 23, 2012