2020
|
Lu, Fei; Maggioni, Mauro; Tang, Sui Learning interaction kernels in stochastic systems of interacting particles from multiple trajectories Journal Article Forthcoming arXiv, Forthcoming. Abstract | Links | BibTeX | Tags: agent-based models, interacting particle systems, inverse problems, Machine learning, statistics, stochastic systems @article{learningStochasticInteracting,
title = {Learning interaction kernels in stochastic systems of interacting particles from multiple trajectories},
author = {Fei Lu and Mauro Maggioni and Sui Tang},
url = {https://arxiv.org/abs/2007.15174},
year = {2020},
date = {2020-07-30},
journal = {arXiv},
abstract = {We consider stochastic systems of interacting particles or agents, with dynamics determined by an interaction kernel which only depends on pairwise distances. We study the problem of inferring this interaction kernel from observations of the positions of the particles, in either continuous or discrete time, along multiple independent trajectories. We introduce a nonparametric inference approach to this inverse problem, based on a regularized maximum likelihood estimator constrained to suitable hypothesis spaces adaptive to data. We show that a coercivity condition enables us to control the condition number of this problem and prove the consistency of our estimator, and that in fact it converges at a near-optimal learning rate, equal to the min-max rate of 1-dimensional non-parametric regression. In particular, this rate is independent of the dimension of the state space, which is typically very high. We also analyze the discretization errors in the case of discrete-time observations, showing that it is of order 1/2 in terms of the time gaps between observations. This term, when large, dominates the sampling error and the approximation error, preventing convergence of the estimator. Finally, we exhibit an efficient parallel algorithm to construct the estimator from data, and we demonstrate the effectiveness of our algorithm with numerical tests on prototype systems including stochastic opinion dynamics and a Lennard-Jones model.},
keywords = {agent-based models, interacting particle systems, inverse problems, Machine learning, statistics, stochastic systems},
pubstate = {forthcoming},
tppubtype = {article}
}
We consider stochastic systems of interacting particles or agents, with dynamics determined by an interaction kernel which only depends on pairwise distances. We study the problem of inferring this interaction kernel from observations of the positions of the particles, in either continuous or discrete time, along multiple independent trajectories. We introduce a nonparametric inference approach to this inverse problem, based on a regularized maximum likelihood estimator constrained to suitable hypothesis spaces adaptive to data. We show that a coercivity condition enables us to control the condition number of this problem and prove the consistency of our estimator, and that in fact it converges at a near-optimal learning rate, equal to the min-max rate of 1-dimensional non-parametric regression. In particular, this rate is independent of the dimension of the state space, which is typically very high. We also analyze the discretization errors in the case of discrete-time observations, showing that it is of order 1/2 in terms of the time gaps between observations. This term, when large, dominates the sampling error and the approximation error, preventing convergence of the estimator. Finally, we exhibit an efficient parallel algorithm to construct the estimator from data, and we demonstrate the effectiveness of our algorithm with numerical tests on prototype systems including stochastic opinion dynamics and a Lennard-Jones model. |
2017
|
Crosskey, Miles C; Maggioni, Mauro ATLAS: A geometric approach to learning high-dimensional stochastic systems near manifolds Journal Article Journal of Multiscale Modeling and Simulation, 15 (1), pp. 110–156, 2017, (arxiv: 1404.0667). Links | BibTeX | Tags: diffusion geometry, Machine learning, Manifold Learning, statistics, stochastic systems @article{CM:ATLAS,
title = {ATLAS: A geometric approach to learning high-dimensional stochastic systems near manifolds},
author = {Miles C Crosskey and Mauro Maggioni},
url = {https://arxiv.org/abs/1404.0667
https://doi.org/10.1137/140970951},
year = {2017},
date = {2017-01-01},
journal = {Journal of Multiscale Modeling and Simulation},
volume = {15},
number = {1},
pages = {110--156},
note = {arxiv: 1404.0667},
keywords = {diffusion geometry, Machine learning, Manifold Learning, statistics, stochastic systems},
pubstate = {published},
tppubtype = {article}
}
|
2013
|
Crosskey, Miles C; Maggioni, Mauro Learning of intrinsically low-dimensional stochastic systems in high-dimensions, I Technical Report 2013, (in preparation). BibTeX | Tags: Manifold Learning, stochastic systems @techreport{CM:MultiscaleDynamicsI,
title = {Learning of intrinsically low-dimensional stochastic systems in high-dimensions, I},
author = {Miles C Crosskey and Mauro Maggioni},
year = {2013},
date = {2013-01-01},
note = {in preparation},
keywords = {Manifold Learning, stochastic systems},
pubstate = {published},
tppubtype = {techreport}
}
|
2011
|
Zheng, W; Rohrdanz, M A; Maggioni, Mauro; Clementi, Cecilia Polymer reversal rate calculated via locally scaled diffusion map Journal Article J. Chem. Phys., (134), pp. 144108, 2011. BibTeX | Tags: diffusion geometry, Machine learning, Manifold Learning, molecular dynamics, stochastic systems @article{ZRMC:PolymerReversal,
title = {Polymer reversal rate calculated via locally scaled diffusion map},
author = {W Zheng and M A Rohrdanz and Mauro Maggioni and Cecilia Clementi},
year = {2011},
date = {2011-01-01},
journal = {J. Chem. Phys.},
number = {134},
pages = {144108},
keywords = {diffusion geometry, Machine learning, Manifold Learning, molecular dynamics, stochastic systems},
pubstate = {published},
tppubtype = {article}
}
|
Rohrdanz, M A; Zheng, W; Maggioni, Mauro; Clementi, Cecilia Determination of reaction coordinates via locally scaled diffusion map Journal Article J. Chem. Phys., (134), pp. 124116, 2011. BibTeX | Tags: diffusion geometry, Machine learning, Manifold Learning, molecular dynamics, stochastic systems @article{RZMC:ReactionCoordinatesLocalScaling,
title = {Determination of reaction coordinates via locally scaled diffusion map},
author = {M A Rohrdanz and W Zheng and Mauro Maggioni and Cecilia Clementi},
year = {2011},
date = {2011-01-01},
journal = {J. Chem. Phys.},
number = {134},
pages = {124116},
keywords = {diffusion geometry, Machine learning, Manifold Learning, molecular dynamics, stochastic systems},
pubstate = {published},
tppubtype = {article}
}
|
2008
|
Coifman, Ronald R; Kevrekidis, Ioannis G; Lafon, Stephane; Maggioni, Mauro; Nadler, Boaz Diffusion Maps, reduction coordinates and low dimensional representation of stochastic systems Journal Article SIAM J.M.M.S., 7 (2), pp. 842–864, 2008. BibTeX | Tags: diffusion geometry, dynamical systems, Laplacian eigenfunctions, Machine learning, model reduction, stochastic systems @article{CKLMN:DiffusionMapsReductionCoordinates,
title = {Diffusion Maps, reduction coordinates and low dimensional representation of stochastic systems},
author = {Ronald R Coifman and Ioannis G Kevrekidis and Stephane Lafon and Mauro Maggioni and Boaz Nadler},
year = {2008},
date = {2008-01-01},
journal = {SIAM J.M.M.S.},
volume = {7},
number = {2},
pages = {842--864},
keywords = {diffusion geometry, dynamical systems, Laplacian eigenfunctions, Machine learning, model reduction, stochastic systems},
pubstate = {published},
tppubtype = {article}
}
|
2007
|
Coifman, Ronald R; Maggioni, Mauro Multiscale Data Analysis with Diffusion Wavelets Journal Article Proc. SIAM Bioinf. Workshop, Minneapolis, 2007. BibTeX | Tags: diffusion geometry, diffusion wavelets, Machine learning, Manifold Learning, multiscale analysis, random walks, spectral graph theory, stochastic systems @article{CM:MsDataDiffWavelets,
title = {Multiscale Data Analysis with Diffusion Wavelets},
author = {Ronald R Coifman and Mauro Maggioni},
year = {2007},
date = {2007-04-01},
journal = {Proc. SIAM Bioinf. Workshop, Minneapolis},
keywords = {diffusion geometry, diffusion wavelets, Machine learning, Manifold Learning, multiscale analysis, random walks, spectral graph theory, stochastic systems},
pubstate = {published},
tppubtype = {article}
}
|
2006
|
Coifman, Ronald R; Maggioni, Mauro Diffusion Wavelets Journal Article Appl. Comp. Harm. Anal., 21 (1), pp. 53–94, 2006, ((Tech. Rep. YALE/DCS/TR-1303, Yale Univ., Sep. 2004)). BibTeX | Tags: diffusion geometry, diffusion wavelets, Machine learning, Manifold Learning, multiscale analysis, random walks, spectral graph theory, stochastic systems @article{CMDiffusionWavelets,
title = {Diffusion Wavelets},
author = {Ronald R Coifman and Mauro Maggioni},
year = {2006},
date = {2006-07-01},
journal = {Appl. Comp. Harm. Anal.},
volume = {21},
number = {1},
pages = {53--94},
note = {(Tech. Rep. YALE/DCS/TR-1303, Yale Univ., Sep. 2004)},
keywords = {diffusion geometry, diffusion wavelets, Machine learning, Manifold Learning, multiscale analysis, random walks, spectral graph theory, stochastic systems},
pubstate = {published},
tppubtype = {article}
}
|
Bremer, James Jr. C; Coifman, Ronald R; Maggioni, Mauro; Szlam, Arthur D Diffusion Wavelet Packets Journal Article Appl. Comp. Harm. Anal., 21 (1), pp. 95–112, 2006, ((Tech. Rep. YALE/DCS/TR-1304, 2004)). BibTeX | Tags: diffusion geometry, Machine learning, Manifold Learning, multiscale analysis, random walks, spectral graph theory, stochastic systems @article{DiffusionWaveletPackets,
title = {Diffusion Wavelet Packets},
author = {James Jr. C Bremer and Ronald R Coifman and Mauro Maggioni and Arthur D Szlam},
year = {2006},
date = {2006-07-01},
journal = {Appl. Comp. Harm. Anal.},
volume = {21},
number = {1},
pages = {95--112},
note = {(Tech. Rep. YALE/DCS/TR-1304, 2004)},
keywords = {diffusion geometry, Machine learning, Manifold Learning, multiscale analysis, random walks, spectral graph theory, stochastic systems},
pubstate = {published},
tppubtype = {article}
}
|
Coifman, Ronald R; Lafon, Stephane; Maggioni, Mauro; Keller, Y; Szlam, A D; Warner, F J; Zucker, S W Geometries of sensor outputs, inference, and information processing Inproceedings Athale, John Zolper; Eds. Intelligent Integrated Microsystems; Ravindra C A (Ed.): Proc. SPIE, pp. 623209, 2006. BibTeX | Tags: diffusion geometry, Laplacian eigenfunctions, Machine learning, Manifold Learning, random walks, spectral graph theory, stochastic systems @inproceedings{CLMKSWZ:GeometrySensorOutputs,
title = {Geometries of sensor outputs, inference, and information processing},
author = {Ronald R Coifman and Stephane Lafon and Mauro Maggioni and Y Keller and A D Szlam and F J Warner and S W Zucker},
editor = {John Zolper; Eds. C Intelligent Integrated Microsystems; Ravindra A. Athale},
year = {2006},
date = {2006-05-01},
booktitle = {Proc. SPIE},
volume = {6232},
pages = {623209},
keywords = {diffusion geometry, Laplacian eigenfunctions, Machine learning, Manifold Learning, random walks, spectral graph theory, stochastic systems},
pubstate = {published},
tppubtype = {inproceedings}
}
|
2005
|
Coifman, Ronald R; Maggioni, Mauro; Zucker, Steven W; Kevrekidis, Ioannis G Geometric diffusions for the analysis of data from sensor networks Journal Article Curr Opin Neurobiol, 15 (5), pp. 576–84, 2005. BibTeX | Tags: diffusion geometry, Laplacian eigenfunctions, Machine learning, Manifold Learning, random walks, spectral graph theory, stochastic systems @article{CMZK:CONB,
title = {Geometric diffusions for the analysis of data from sensor networks},
author = {Ronald R Coifman and Mauro Maggioni and Steven W Zucker and Ioannis G Kevrekidis},
year = {2005},
date = {2005-10-01},
journal = {Curr Opin Neurobiol},
volume = {15},
number = {5},
pages = {576--84},
keywords = {diffusion geometry, Laplacian eigenfunctions, Machine learning, Manifold Learning, random walks, spectral graph theory, stochastic systems},
pubstate = {published},
tppubtype = {article}
}
|
Coifman, Ronald R; Lafon, Stephane; Lee, Ann B; Maggioni, Mauro; Nadler, B; Warner, Frederick; Zucker, Steven W Geometric diffusions as a tool for harmonic analysis and structure definition of data: Diffusion maps Journal Article Proceedings of the National Academy of Sciences of the United States of America, 102 (21), pp. 7426-7431, 2005. BibTeX | Tags: diffusion geometry, Machine learning, Manifold Learning, random walks, spectral graph theory, stochastic systems @article{DiffusionPNAS,
title = {Geometric diffusions as a tool for harmonic analysis and structure definition of data: Diffusion maps},
author = {Ronald R Coifman and Stephane Lafon and Ann B Lee and Mauro Maggioni and B Nadler and Frederick Warner and Steven W Zucker},
year = {2005},
date = {2005-01-01},
journal = {Proceedings of the National Academy of Sciences of the United States of America},
volume = {102},
number = {21},
pages = {7426-7431},
keywords = {diffusion geometry, Machine learning, Manifold Learning, random walks, spectral graph theory, stochastic systems},
pubstate = {published},
tppubtype = {article}
}
|
Coifman, Ronald R; Lafon, S; Lee, A B; Maggioni, Mauro; Nadler, B; Warner, Frederick; Zucker, Steven W Geometric diffusions as a tool for harmonic analysis and structure definition of data: Multiscale methods Journal Article Proceedings of the National Academy of Sciences of the United States of America, 102 (21), pp. 7432–7438, 2005. BibTeX | Tags: diffusion geometry, Machine learning, Manifold Learning, multiscale analysis, random walks, spectral graph theory, stochastic systems @article{DiffusionPNAS2,
title = {Geometric diffusions as a tool for harmonic analysis and structure definition of data: Multiscale methods},
author = {Ronald R Coifman and S Lafon and A B Lee and Mauro Maggioni and B Nadler and Frederick Warner and Steven W Zucker},
year = {2005},
date = {2005-01-01},
journal = {Proceedings of the National Academy of Sciences of the United States of America},
volume = {102},
number = {21},
pages = {7432--7438},
keywords = {diffusion geometry, Machine learning, Manifold Learning, multiscale analysis, random walks, spectral graph theory, stochastic systems},
pubstate = {published},
tppubtype = {article}
}
|
2004
|
Coifman, Ronald R; Maggioni, Mauro Multiresolution Analysis associated to diffusion semigroups: construction and fast algorithms Technical Report Dept. Comp. Sci., Yale University (YALE/DCS/TR-1289), 2004. BibTeX | Tags: diffusion geometry, Machine learning, Manifold Learning, multiscale analysis, random walks, spectral graph theory, stochastic systems @techreport{CMTech,
title = {Multiresolution Analysis associated to diffusion semigroups: construction and fast algorithms},
author = {Ronald R Coifman and Mauro Maggioni},
year = {2004},
date = {2004-05-01},
number = {YALE/DCS/TR-1289},
institution = {Dept. Comp. Sci., Yale University},
keywords = {diffusion geometry, Machine learning, Manifold Learning, multiscale analysis, random walks, spectral graph theory, stochastic systems},
pubstate = {published},
tppubtype = {techreport}
}
|