Publications2021From perspective maps to epigraphical projectionM. P. Friedlander, A. Goodwin, T. HoheiselarXiv:2102.06809, 2021Fast convergence of the stochastic subgradient method under interpolationH. Fang, Z. Fan, M. P. FriedlanderIntern. Conf. Learning Representations (ICLR), 20212020Atomic decomposition via polar alignment: the geometry of structured optimizationZ. Fan, H. Jeong, Y. Sun, M. P. FriedlanderFoundations and Trends in Optimization, 3(4):280–366, 2020Polar deconvolution of mixed signalsZ. Fan, H. Jeong, B. Joshi, M. P. FriedlanderarXiv:2010.10508, 2020Implementing a smooth exact penalty function for equality-constrained nonlinear optimizationR. Estrin, M. P. Friedlander, D. Orban, M. A. SaundersSIAM J. Sci. Comput., 42(3), A1809–A1835, 2020Implementing a smooth exact penalty function for general constrained nonlinear optimizationR. Estrin, M. P. Friedlander, D. Orban, M. A. SaundersSIAM J. Sci. Comput., 42(3), A1836–A1859, 2020A perturbation view of level-set methods for convex optimizationR. Estrin, M. P. FriedlanderOptimization Letters, 2020Online mirror descent and dual averaging: keeping pace in the dynamic caseH. Fang, N J. A. Harvey, V. S. Portella, M. P. FriedlanderIntern. Conf. Machine Learning (ICML), 2020Greed meets sparsity: understanding and improving greedy coordinate descent for sparse optimizationH. Fang, Z. Fan, Y. Sun, M. P. FriedlanderIntern. Conf. Artificial Intelligence and Statistics (AISTATS), 20202019Bundle methods for dual atomic pursuitZ. Fan, Y. Sun, M. P. FriedlanderAsilomar Conf. Signals, Systems, Computers (ACSSC), 2019One-shot atomic detectionY. Sun, M. P. FriedlanderIEEE Intern. Workshop Comput. Adv. Multi-Sensor Adaptive Proc. (CAMSAP), 2019Fast training for large-scale one-versus-all linear classifiers using tree-structured initializationH. Fang, M. Cheng, C.-J. Hsieh, M. P. FriedlanderSIAM Intern. Conf. Data Mining (SDM), 2019Polar convolutionM. P. Friedlander, I. Macêdo, T.K. PongSIAM J. Optimization, 29(2):1366–1391, 2019Smooth structured prediction using quantum and classical gibbs samplersB. Sepehry, E. Iranmanesh, M. P. Friedlander, P. RonaghAdiabatic Quantum Computing Conference (AQC), arXiv:1809.04091, 20192018Foundations of gauge and perspective dualityA. Y. Aravkin, J. V. Burke, D. Drusvyatskiy, M. P. Friedlander, K. MacPheeSIAM J. Optimization, 28(3):2406–2434, 2018Level-set methods for convex optimizationA. Y. Aravkin, J. V. Burke, D. Drusvyatskiy, M. P. Friedlander, S. RoyMathematical Programming, 174(1-2):359–390, 20182017Efficient evaluation of scaled proximal operatorsM. P. Friedlander, G. GohElectronic Trans. Numerical Analysis, 46:1–22, 20172016Satisfying real-world goals with dataset constraintsG. Goh, A. Cotter, M. Gupta, M. P. FriedlanderAdvances in Neural Information Processing Systems 29 (NIPS), 2016Social resistanceM. P. Friedlander, N. Krislock, T. K. PongIEEE Comput. Sci. Eng., 8(2):98-103; reprinted in Computing Edge, 2016Low-rank spectral optimization via gauge dualityM. P. Friedlander, I. MacêdoSIAM J. Scientific Computing, 38(3):A1616–A1638, 20162015Coordinate descent converges faster with the Gauss-Southwell rule than random selectionJ. Nutini, M. Schmidt, I. H. Laradji, M. P. Friedlander, H. KoepkeIntern. Conf. Machine Learning (ICML), 20152014Gauge optimization and dualityM. P. Friedlander, I. Macêdo, T. K. PongSIAM J. Optimization, 24(4):1999–2022, 20142013Variational properties of value functionsA. Aravkin, J. V. Burke, M. P. FriedlanderSIAM J. Optimization, 23(3):1689–1717, 2013Fast dual variational inference for non-conjugate latent gaussian modelsM. E. Khan, A. Y. Aravkin, M. P. Friedlander, M. SeegerIntern. Conf. Machine Learning (ICML), 2013Tail bounds for stochastic approximationM. P. Friedlander, G. GoharXiv:1304.5586, 20132012Hybrid deterministic-stochastic methods for data fittingM. P. Friedlander, M. SchmidtSIAM J. Scientific Computing, 34(3):A1380–A1405, 2012A primal-dual regularized interior-point method for convex quadratic programsM. P. Friedlander, D. OrbanMathematical Programming Computation, 4(1):71–107, 2012Fighting the curse of dimensionality: compressive sensing in exploration seismologyF. J. Herrmann, M. P. Friedlander, Ö. YılmazIEEE Signal Processing Magazine, 29(3):88–100, 2012Robust inversion via semistochastic dimensionality reductionA. Aravkin, M. P. Friedlander, T. van LeeuwenIntern. Conf. Acoustics, Speech, and Signal Processing (ICASSP), 2012Robust inversion, dimensionality reduction, and randomized samplingA. Aravkin, M. P. Friedlander, F. Herrmann, T. van LeeuwenMathematical Programming, 134(1):101–125, 2012, 20122011Sparse optimization with least-squares constraintsE. van den Berg, M. P. FriedlanderSIAM J. Optimization, 21(4):1201–1229, 2011Recovering compressively sampled signals using partial support informationM. P. Friedlander, H. Mansour, R. Saab, Ö. YılmazIEEE Trans. Info. Theory, 58(2):1122–1134, 2011Theoretical and empirical results for recovery from multiple measurementsE. van den Berg, M. P. FriedlanderIEEE Trans. Info. Theory, 56(5):2516–2527, 20112009Optimizing costly functions with simple constraints: a limited-memory projected quasi-Newton algorithmM. Schmidt, E. van den Berg, M. P. Friedlander, K. MurphyIntern. Conf. Artificial Intelligence and Statistics (AISTATS), 2009Sparco: a testing framework for sparse reconstructionE. van den Berg, M. P. Friedlander, G. Hennenfent, F. Herrmann, R. Saab, Ö. YılmazACM Trans. Math. Software, 35(4):1–16, 20092008Probing the Pareto frontier for basis pursuit solutionsE. van den Berg, M. P. FriedlanderSIAM J. Scientific Computing, 31(2):890–912, 2008Computing nonnegative tensor factorizationsM. P. Friedlander, K. HatzOptimization Methods and Software, 23(4):631–647, 2008New insights into one-norm solvers from the Pareto curveG. Hennenfent, E. van den Berg, M. P. Friedlander, F. HerrmannGeophysics, 73(4):A23–A26, 2008Group sparsity via linear-time projectionE. van den Berg, M. Schmidt, M. P. Friedlander, K. MurphyTech. Rep. TR-2008-09, Dept of Computer Science, UBC, 2008Global and finite termination of a two-phase augmented Lagrangian filter method for general quadratic programsM. P. Friedlander, S. LeyfferSIAM J. Scientific Computing, 30(4):1706–1726, 20082007Discussion: The Dantzig selector: Statistical estimation when p is much larger than nM. P. Friedlander, M. A. SaundersAnnals of Statistics, 35(6):2385–2391, 2007Exact regularization of convex programsM. P. Friedlander, P. TsengSIAM J. Optimization, 18(4):1326–1350, 2007A filter active-set trust-region methodM. P. Friedlander, N. I. M. Gould, S. Leyffer, T. S. MunsonPreprint ANL/MCS-P1456-0907, Argonne National Laboratory, 2007In pursuit of a rootE. van den Berg, M. P. FriedlanderTech. Rep. TR-2007-19, Dept. of Computer Science, UBC, 2007Diffuse optical fluorescence tomography using time-resolved data acquired in transmissionF. Leblond, S. Fortier, M. P. FriedlanderMultimodal Biomedical Imaging II, vol. 6431. Proc. Intern. Society Optimal Imaging, 20072006On minimizing distortion and relative entropyM. P. Friedlander, M. R. GuptaIEEE Trans. Info. Theory, 52(1):238–245, 20062005A two-sided relaxation scheme for mathematical programs with equilibrium constraintsV. Demiguel, M. P. Friedlander, F. J. Nogales, S. ScholtesSIAM J. Optimization, 16(1):587–609, 2005A globally convergent linearly constrained Lagrangian method for nonlinear optimizationM. P. Friedlander, M. A. SaundersSIAM J. Optimization 15(3):863–897, 20052000Maximum entropy classification applied to speechM. R. Gupta, M. P. Friedlander, R. M. GrayAsilomar Conf. Signals, Systems, Computers (ACSSC), vol. 2, 1480–1483, 2000Selected TalksPolar deconvolution of mixed signals, One-World Optimization Seminar. October, 26, 2020Polar Duality in three liftings, McGill University, Montreal. November 28, 2017Algorithms for sparse optimization, Google Research, Mountain View. May 19, 2015Robust inversion and randomized sampling, International Symposium on Mathematical Programming (ISMP), Berlin. August 23, 2012