Department of Mathematics
 Search | Help | Login | pdf version | printable version

Math @ Duke





.......................

.......................


Mauro Maggioni, Professor of Mathematics and Electrical and Computer Engineering and Computer Science

Mauro Maggioni

I am interested in novel constructions inspired by classical harmonic analysis that allow to analyse the geometry of manifolds and graphs and functions on such structures. These constructions are motivated by several important applications across many fields. In many situations we are confronted with large amounts of apparently unstructured high-dimensional data. I find fascinating to study the intrinsic geometry of such data, and exploiting in order to study, explore, visualize, characterize statistical properties of the data. Oftentimes such data is modeled as a manifold (or something "close to a manifold") or a graph, and functions on these spaces need to approximated or "learned" from the data and experiments on the data. For example each data point could be a document, a graph associated with the documents could be given by for example hyperlinks, or by similarity of word frequencies, and a function on the set of documents would be how interesting I personally score a document. One may wish to learn how to predict how much I would score documents I have not seen yet. This can be cast as an approximation problem on the graph of documents, and it turns out that one can generalize Euclidean-type approximation techniques (in particular multiscale regression techniques) to tackle this problem. An application of the above techniques that I find particularly interesting is Markov Decision Processes and Reinforcement Learning, where the problem of learning a behaviour from experience is cast in a rather general optimization and learning framework that involves approximations of functions and operators on graphs and manifolds. I am also interested in imaging, in particular I am working on novel classes of nonlinear denoising algorithms, based on diffusion processes on graphs of features built from images. Another interest is in the geometry of multiscale dynamical systems, and the construction of algorithms for the empirical construction of approximate equations for such systems. I also work on hyperspectral imaging, in particular in building automatic classifiers for discriminating normal from cancerous biopsies, for automated diagnostics and pathology.

Contact Info:
Office Location:  117 Physics Bldg, Office #293, Durham, NC 27708
Office Phone:  (919) 660-2825
Email Address: send me a message
Web Page:  http://www.math.duke.edu/~mauro

Teaching (Spring 2016):

  • MATH 431.01, ADVANCED CALCULUS I Synopsis
    Physics 235, MW 10:05 AM-11:20 AM
Education:

Ph.D.Washington University2002
M.S.Washington University2000
Laurea in MatematicaUniversita' degli Studi di Milano, Italy1999
Specialties:

Applied Math
Analysis
Probability
Research Interests: Harmonic analysis, with applications to statistical analysis of high-dimensional data, machine learning, imaging.

Current projects: Multiscale analysis on graphs and manifolds, Nonlinear image denoising, Compressed imaging and hyperspectral imaging, Supervised and semisupervised learning on manifolds and graphs, Universal mappings via the eigenfunctions of the Laplacian, Perturbation of eigenfunctions of the Laplacian on graphs, Multiscale manifolds methods for Markov Decision Processes

I am interested in novel constructions inspired by classical harmonic analysis that allow to analyse the geometry of manifolds and graphs and functions on such structures. These constructions are motivated by several important applications across many fields. In many situations we are confronted with large amounts of apparently unstructured high-dimensional data. I find fascinating to study the intrinsic geometry of such data, and exploiting in order to study, explore, visualize, characterize statistical properties of the data. Oftentimes such data is modeled as a manifold (or something "close to a manifold") or a graph, and functions on these spaces need to approximated or "learned" from the data and experiments on the data. For example each data point could be a document, a graph associated with the documents could be given by for example hyperlinks, or by similarity of word frequencies, and a function on the set of documents would be how interesting I personally score a document. One may wish to learn how to predict how much I would score documents I have not seen yet. This can be cast as an approximation problem on the graph of documents, and it turns out that one can generalize Euclidean-type approximation techniques (in particular multiscale regression techniques) to tackle this problem. An application of the above techniques that I find particularly interesting is Markov Decision Processes and Reinforcement Learning, where the problem of learning a behaviour from experience is cast in a rather general optimization and learning framework that involves approximations of functions and operators on graphs and manifolds. I am also interested in imaging, in particular I am working on novel classes of nonlinear denoising algorithms, based on diffusion processes on graphs of features built from images. Another interest is in the geometry of multiscale dynamical systems, and the construction of algorithms for the empirical construction of approximate equations for such systems. I also work on hyperspectral imaging, in particular in building automatic classifiers for discriminating normal from cancerous biopsies, for automated diagnostics and pathology.

Areas of Interest:

Harmonic analysis
Multiscale analysis
Markov decision processes
Machine learning
High-dimensional data analysis
Stochastic dynamical systems
Signal processing
Imaging (e.g. hyperspectral)
Geometric measure theory

Keywords:

Algorithms • Amino Acids, Branched-Chain • Animals • Anxiety • Base Sequence • Behavior, Animal • Brain • Chemistry, Physical • Chromosome Mapping • Diet • Diet, High-Fat • Dietary Sucrose • Diffusion • Exploratory Behavior • Face • Harmonic • Hyperspectral • Imaging • Imaging, Three-Dimensional • Kinetics • Kynurenic Acid • Laplacian • Male • Models, Neurological • Models, Theoretical • Molecular Conformation • Molecular Dynamics Simulation • Mood Disorders • Multiscale • Multiscale Dynamical systems • Nanopores • Neural Networks (Computer) • Neurons • Obesity • Polymers • Rats • Rats, Wistar • Serotonin • Spectral graph theory • Thermodynamics • Time Factors • Tryptophan • Weight Gain

Curriculum Vitae
Current Ph.D. Students   (Former Students)

Postdocs Mentored

  • James Murphy (2015/12-present)  
  • Stefano Vigogna (January, 2015 - present)  
  • Wenjing Liao (August, 2013 - present)  
  • David Lawlor (2012 - 2015)  
  • Joshua Vogelstein (2012 - 2014)  
  • Samuel Gerber (2012 - 2015)  
  • Grace Yi Wang (September, 2012 - present)  
  • Nate Strawn (2011 - 2014)  
  • Mark Iwen (2010 - 2013)  
  • Guangliang Chen (2009 - 2012)  
  • Jake Bouvrie (2009 - 2012)  
  • Yoon-Mo Jung (2007 - 2009)  
Undergraduate Research Supervised

  • Jason Lee (May, 2009 - May, 2010)  
Representative Publications   (More Publications)

  1. WK Allard, G Chen and M Maggioni, Multiscale Geometric Methods for Data Sets II: Geometric Wavelets, CoRR, vol. abs/1105.4924 no. 3 (2012)
  2. MA Iwen and M Maggioni, Approximation of Points on Low-Dimensional Manifolds Via Random Linear Projections, arxiv, vol. 2 (February, 2013) [1204.3337v1], [doi]  [abs]
  3. M. Maggioni, Geometric Estimation of Probability Measures in High-Dimensions, Proc. IEEE Asilomar Conference (November, 2013)
  4. S Gerber and M Maggioni, Multiscale dictionaries, transforms, and learning in high-dimensions, WAVELETS AND SPARSITY XV, vol. 8858 (2013), ISSN 0277-786X [Gateway.cgi], [doi]
  5. M Maggioni, GL Davis, FJ Warner, FB Geshwind, AC Coppi, RA DeVerse and RR Coifman, Hyperspectral microscopic analysis of normal, benign and carcinoma microarray tissue sections, edited by Robert R. Alfano and Alvin Katz, Progress in Biomedical Optics and Imaging - Proceedings of SPIE, vol. 6091 no. 1 (2006), pp. 60910I, SPIE, ISSN 1605-7422 [1], [doi]  [abs]
  6. RR Coifman and M Maggioni, Diffusion wavelets, Applied and Computational Harmonic Analysis, vol. 21 no. 1 (2006), pp. 53-94, ISSN 1063-5203 [doi]  [abs]
  7. J Bouvrie and M Maggioni, Efficient solution of Markov decision problems with multiscale representations, 2012 50th Annual Allerton Conference on Communication, Control, and Computing, Allerton 2012 (2012), pp. 474-481 [doi]  [abs]
  8. Sridhar Mahadevan and Mauro Maggioni, Proto-value Functions: A Spectral Framework for Solving Markov Decision Processes, submitted (2006)
  9. JV Bouvrie and M Maggioni, Geometric multiscale reduction for autonomous and controlled nonlinear systems., in Proc. IEEE Conference on Decision and Control (CDC), CDC (2012), pp. 4320-4327, IEEE, ISBN 978-1-4673-2065-8 [mostRecentIssue.jsp], [doi]
  10. Mauro Maggioni and Sridhar Mahadevan, Multiscale Diffusion Bases for Policy Iteration in Markov Decision Processes, submitted (Submitted, 2006)
  11. G. Chen, A.V. Little, M. Maggioni, L. Rosasco, Some recent advances in multiscale geometric analysis of point clouds, in Wavelets and Multiscale Analysis: Theory and Applications (March, 2011), Springer
  12. G. Chen, M. Maggioni, Multiscale Analysis of Plane Arrangements, in Proc. C.V.P.R. (2011)
  13. MA Rohrdanz, W Zheng, M Maggioni and C Clementi, Determination of reaction coordinates via locally scaled diffusion map., J Chem Phys, vol. 134 no. 12 (2011), pp. 124116 [21456654], [doi]  [abs]
  14. G. Chen, M. Maggioni, Multiscale Geometric Dictionaries for point-cloud data, Proc. SampTA 2011 (2011)
  15. W Zheng, MA Rohrdanz, M Maggioni and C Clementi, Polymer reversal rate calculated via locally scaled diffusion map., J Chem Phys, vol. 134 no. 14 (2011), pp. 144109 [21495744], [doi]  [abs]
  16. G Chen and M Maggioni, Multiscale geometric wavelets for the analysis of point clouds, 2010 44th Annual Conference on Information Sciences and Systems, CISS 2010 (February, 2010) [doi]  [abs]
  17. PW Jones, M Maggioni and R Schul, Universal local parametrizations via heat kernels and eigenfunctions of the Laplacian, Annales Academiae Scientiarum Fennicae Mathematica, vol. 35 no. 1 (January, 2010), pp. 131-174, ISSN 1239-629X [0709.1975v4], [doi]  [abs]
  18. J. Lee, M. Maggioni, Multiscale Analysis of Time Series of Graphs, Proc. SampTA 2011 (2010)
  19. PW Jones, M Maggioni and R Schul, Manifold parametrizations by eigenfunctions of the Laplacian and heat kernels., Proc Natl Acad Sci U S A, vol. 105 no. 6 (2008), pp. 1803-1808 [18258744], [doi]  [abs]
  20. RR Coifman, S Lafon, IG Kevrekidis, M Maggioni and B Nadler, Diffusion maps, reduction coordinates, and low dimensional representation of stochastic systems, Multiscale Modeling and Simulation, vol. 7 no. 2 (2008), pp. 842-864, ISSN 1540-3459 [doi]  [abs]
  21. S Mahadevan and M Maggioni, Proto-value Functions: A Laplacian Framework for Learning Representation and Control, Journ. Mach. Learn. Res. no. 8 (September, 2007)
  22. G. Chen, A.V. Little, M. Maggioni, Multi-Resolution Geometric Analysis for Data in High Dimensions, in Excursions in Harmonic Analysis, vol. 1 (2013), Birkha├╝ser Boston, ISBN 978-0-8176-8375-7 [doi]
  23. S Mahadevan and M Maggioni, Value function approximation with diffusion wavelets and Laplacian eigenfunctions, in University of Massachusetts, Department of Computer Science Technical Report TR-2005-38; Proc. NIPS 2005, Advances in Neural Information Processing Systems (2005), pp. 843-850, ISSN 1049-5258  [abs]
  24. NH Katz, E Krop and M Maggioni, On the box problem, Math. Research Letters, vol. 4 (2002), pp. 515-519
  25. RR Coifman, S Lafon, AB Lee, M Maggioni, B Nadler, F Warner and SW Zucker, Geometric diffusions as a tool for harmonic analysis and structure definition of data: multiscale methods., Proc Natl Acad Sci U S A, vol. 102 no. 21 (May, 2005), pp. 7432-7437, ISSN 0027-8424 [15899969], [doi]  [abs]
  26. RR Coifman, S Lafon, AB Lee, M Maggioni, B Nadler, F Warner and SW Zucker, Geometric diffusions as a tool for harmonic analysis and structure definition of data: diffusion maps., Proc Natl Acad Sci U S A, vol. 102 no. 21 (May, 2005), pp. 7426-7431, ISSN 0027-8424 [15899970], [doi]  [abs]
  27. J. Bouvrie, M. Maggioni, Multiscale Markov Decision Problems: Compression, Solution, and Transfer Learning (Submitted, 2012) [1212.1143]
Recent Grant Support

  • BIGDATA:Collaborative Research:F:From Data Geometries to Information Networks, National Science Foundation, 1546392, 2016/01-2019/12.      
  • Statistical learning for high-dimensional stochastic dynamical systems, National Science Foundation, 1522651, 2015/09-2018/08.      
  • Information, Approximation, and Fast Algorithms for Data in High Dimensions, Air Force Office of Scientific Research, 2013/12-2016/12.      
  • Geometric Approximation and Estimation of Probability Measures in High Dimensions, Air Force Office of Scientific Research, FA9550-14-1-0033, 2013/12-2016/12.      
  • Collaborative Research: SI2-CHE-ExTASY Extensible Tools for Advanced Sampling and AnalYsis, National Science Foundation, CHE-1265920, 2013/09-2016/08.      
  • ATD: Online Multiscale Algorithms for Geometric Density Estimation in High-Dimensions and Persistent Homology of Data fo, National Science Foundation, DMS-1222567, 2012/09-2016/08.      
  • ATD: Online Multiscale Algorithms for Geometric Density Estimation in High-Dimensions and Persistent Homology of Data fo, National Science Foundation, DMS-1222567, 2012/09-2016/08.      
  • Multiscale Analysis of Dynamic Graphs, Office of Naval Research, N00014-12-1-0601-01, 2012/08-2016/07.      
  • Multiscale Analysis of Dynamic Graphs, Office of Naval Research, N00014-12-1-0601-01, 2012/08-2016/07.      
  • A Rigorous Statistical Framework for the Mathematics of Sensing, Exploitation and Execution, Defense Advanced Research Projects Agency, 2011/11-2014/11.      
  • Mathematical Foundations of Multiscale Graph Representations and Interactive Learning, National Science Foundation, 2008/05-2014/07.      
  • CAREER, National Science Foundation, DMS-0847388, 2009/07-2014/06.      
  • Knowledge Enhanced Exapixel Photography, Defense Advanced Research Projects Agency, N66001-11-1-4002, 2010/11-2013/11.      
  • Collaborative Proposal, National Science Foundation, NSF-CHE-0835712, 2008/10-2013/09.      
  • NeTS: Small: Collaborative Research:, National Science Foundation, IIS-0916855, 2009/09-2013/08.      
  • Multiscale geometry for the analysis of high dimensional datasets, Washington State University, 113054 G002745, 2010/05-2013/05.      
Conferences Organized

  • SAMSI-FODAVA Workshop on Interactive Visualization and Analysis of Massive Data. December 10, 2012, December 10, 2012  
  • Symposium of Knowledge Extraction at A.M.S. nat. meeing 2010. January 13, 2010, January 13, 2010  
  • Large Data Workshop, C.T.M.S., Duke. November 13, 2009, November 13, 2009  
  • A.A.A.I. workshop on manifold learning. November 5, 2009, November 5, 2009  
  • Organizer : Internet Multi-Resolution Analysis: Foundations, Applications and Practice. September 2008 - December 2008, Organizer, September, 2008 - December, 2008  
  • Workshop on Eigenfunctions of the Laplacian, ICIAM 2007, July 18, 2007  
  • Document Space, Organizer, January, 2006  

 

dept@math.duke.edu
ph: 919.660.2800
fax: 919.660.2821

Mathematics Department
Duke University, Box 90320
Durham, NC 27708-0320