when models and data disagree sparse resolutions to
play

When models and data disagree: sparse resolutions to inconsistent - PowerPoint PPT Presentation

Oct 05, 2022 •13 likes •443 views

When models and data disagree: sparse resolutions to inconsistent datasets Arun Hegde Wenyu Li Jim Oreluk Andrew Packard Michael Frenklach This project is supported by the U.S. Department of Energy, National Nuclear Security Administration,

  1. When models and data disagree: sparse resolutions to inconsistent datasets Arun Hegde Wenyu Li Jim Oreluk Andrew Packard Michael Frenklach This project is supported by the U.S. Department of Energy, National Nuclear Security Administration, under Award Number DE-NA0002375 SIAM UQ18 APRIL 16-19, 2018

  2. Overview • Bound-to-Bound Data Collaboration (B2BDC) • models + data = dataset (model-data system) • Dataset Consistency – agreement between models and data • scalar consistency measure • vector consistency measure • Dataset examples • GRI-Mech 3.0 • DLR-SynG • Summary SIAM UQ18 APRIL 16-19, 2018

  3. Bound-to-Bound Data Collaboration UQ as constrained optimization: parameters constrained by models and data Models Data Dataset • Prior knowledge on “True” QOI models uncertain parameters Surrogate QOI models • QOI measurements (w/ uncertainty) Fitting error Feasible set ─ parameters for which the models and data agree. Prediction bounds the range of a model subject to the feasible set SIAM UQ18 APRIL 16-19, 2018

  4. Consistency as Model Validation • A dataset is consistent if it is feasible – Parameters exist for which model predictions match the experiments QOI space Parameter space • Consistency analysis provides measures of validation SIAM UQ18 APRIL 16-19, 2018

  5. Quantifying Consistency Scalar Consistency Measure (SCM)* Q: Does there exist a parameter vector for which the models and data agree, within uncertainty? A: Compute the scalar consistency measure ( SCM ) QOI space *Feeley, R.; Seiler, P.; Packard, A.; Frenklach., M.; J. Phys. Chem. A. 2004, 108 , 9573. SIAM UQ18 APRIL 16-19, 2018

  6. Quantifying Consistency Scalar Consistency Measure (SCM)* Q: Does there exist a parameter vector for which the models and data agree, within uncertainty? A: Compute the scalar consistency measure ( SCM ) If consistent , go to prediction *Feeley, R.; Seiler, P.; Packard, A.; Frenklach., M.; J. Phys. Chem. A. 2004, 108 , 9573. SIAM UQ18 APRIL 16-19, 2018

  7. Quantifying Consistency Scalar Consistency Measure (SCM)* Q: Does there exist a parameter vector for which the models and data agree, within uncertainty? A: Compute the scalar consistency measure ( SCM ) If inconsistent , … ??? *Feeley, R.; Seiler, P.; Packard, A.; Frenklach., M.; J. Phys. Chem. A. 2004, 108 , 9573. SIAM UQ18 APRIL 16-19, 2018

  8. Quantifying Consistency Scalar Consistency Measure (SCM)* Q: Does there exist a parameter vector for which the models and data agree, within uncertainty? A: Compute the scalar consistency measure ( SCM ) • Inconsistency  models and data disagree • Follow-up questions: • What are the sources of inconsistency? • Where do we begin to look? *Feeley, R.; Seiler, P.; Packard, A.; Frenklach., M.; J. Phys. Chem. A. 2004, 108 , 9573. SIAM UQ18 APRIL 16-19, 2018

  9. Quantifying Consistency Scalar Consistency Measure (SCM) Q: Does there exist a parameter vector for which the models and data agree, within uncertainty? A: Compute the scalar consistency measure ( SCM ) Lagrange multipliers from dual form Local: Sensitivities Global: Feeley, R.; Seiler, P.; Packard, A.; Frenklach., M.; J. Phys. Chem. A. 2004, 108 , 9573. SIAM UQ18 APRIL 16-19, 2018

  10. Quantifying Consistency Scalar Consistency Measure (SCM) Q: Does there exist a parameter vector for which the models and data agree, within uncertainty? A: Compute the scalar consistency measure ( SCM ) Q: New question: What is the fewest number of constraint relaxations required to render the dataset consistent? SIAM UQ18 APRIL 16-19, 2018

  11. Vector Consistency Scalar Consistency Measure (SCM) Q: Does there exist a parameter vector for which the models and data agree, within uncertainty? A: Compute the scalar consistency measure ( SCM ) If inconsistent, compute the Vector Consistency Measure (VCM) vector consistency measure ( VCM ) • Offers detailed analysis of inconsistency by allowing independent relaxations. • Can be used to flag constraints contributing to inconsistency SIAM UQ18 APRIL 16-19, 2018

  12. Vector Consistency Scalar Consistency Measure (SCM) Q: Does there exist a parameter vector for which the models and data agree, within uncertainty? A: Compute the scalar consistency measure ( SCM ) If inconsistent, compute the Vector Consistency Measure (VCM) vector consistency measure ( VCM ) heuristic for fewest # of nonzeros (sparsity) • Offers detailed analysis of inconsistency by allowing independent relaxations. • Can be used to flag constraints contributing to inconsistency SIAM UQ18 APRIL 16-19, 2018

  13. Examples * * Hegde, A.; Li, W.; Oreluk, J.; Packard, A.; Frenklach, M., SIAM/ASA J. Uncert. Quantif., 2018, 6(2), 429-456. SIAM UQ18 APRIL 16-19, 2018

  14. Example 1: GRI-Mech 3.0 GRI-Mech 3.0 dataset ( 77 QOIs, 102 uncertain parameters ) for natural gas combustion. Scalar Consistency Vector Consistency • Procedure: apply SCM, use sensitivities to flag problems. SIAM UQ18 APRIL 16-19, 2018

  15. Example 1: GRI-Mech 3.0 GRI-Mech 3.0 dataset ( 77 QOIs, 102 uncertain parameters ) for natural gas combustion. Scalar Consistency Vector Consistency • Procedure: apply SCM, use sensitivities to flag problems. • SCM < 0. Analyze ranked sensitivities SIAM UQ18 APRIL 16-19, 2018

  16. Example 1: GRI-Mech 3.0 GRI-Mech 3.0 dataset ( 77 QOIs, 102 uncertain parameters ) for natural gas combustion. Scalar Consistency Vector Consistency • Procedure: apply SCM, use sensitivities to identify problems. • SCM < 0. Analyze ranked sensitivities SIAM UQ18 APRIL 16-19, 2018

  17. Example 1: GRI-Mech 3.0 GRI-Mech 3.0 dataset ( 77 QOIs, 102 uncertain parameters ) for natural gas combustion. Scalar Consistency Vector Consistency • • Procedure: apply SCM, use Compute VCM. sensitivities to identify problems. • SCM < 0. Analyze ranked sensitivities SIAM UQ18 APRIL 16-19, 2018

  18. Example 1: GRI-Mech 3.0 GRI-Mech 3.0 dataset ( 77 QOIs, 102 uncertain parameters ) for natural gas combustion. Scalar Consistency Vector Consistency • • Procedure: apply SCM, use Compute VCM. sensitivities to flag problems. • Two QOIs relaxed (same as in SCM), • dataset consistent. SCM < 0. Analyze ranked sensitivities • SCM > 0. Two QOIs removed, dataset consistent. SIAM UQ18 APRIL 16-19, 2018

  19. Example 1: GRI-Mech 3.0 GRI-Mech 3.0 dataset ( 77 QOIs, 102 uncertain parameters ) for natural gas combustion. Scalar Consistency Vector Consistency • • Procedure: apply SCM, use Compute VCM. sensitivities to flag problems. • Two QOIs relaxed (same as in SCM), • dataset consistent. SCM < 0. Analyze ranked sensitivities • SCM > 0. 2 QOIs removed, dataset consistent. Rapid and interpretable Rapid and interpretable resolution of inconsistency resolution of inconsistency SIAM UQ18 APRIL 16-19, 2018

  20. Example 2: DLR-SynG DLR-SynG dataset ( 159 QOIs, 55 uncertain parameters ) for syngas combustion developed at DLR*. Scalar Consistency Vector Consistency * Slavinskaya, N.; et al. Energy & Fuels. 2017, 31, pp 2274 – 2297 SIAM UQ18 APRIL 16-19, 2018

  21. Example 2: DLR-SynG DLR-SynG dataset ( 159 QOIs, 55 uncertain parameters ) for syngas combustion developed at DLR. Scalar Consistency Vector Consistency • SCM < 0. Analyze ranked sensitivities. SIAM UQ18 APRIL 16-19, 2018

  22. Example 2: DLR-SynG DLR-SynG dataset ( 159 QOIs, 55 uncertain parameters ) for syngas combustion developed at DLR. Scalar Consistency Vector Consistency • SCM < 0. Analyze ranked sensitivities. Set aside the top most sensitive QOI Set aside the top two most sensitive QOIs Set aside the top n most sensitive QOIs Set aside the second most sensitive QOI (counter intuitive) . . . SIAM UQ18 APRIL 16-19, 2018

  23. Example 2: DLR-SynG DLR-SynG dataset ( 159 QOIs, 55 uncertain parameters ) for syngas combustion developed at DLR. Scalar Consistency Vector Consistency • SCM < 0. Analyze ranked sensitivities. Set aside the top most sensitive QOI Set aside the top two most sensitive QOIs Set aside the top n most sensitive QOIs Set aside the second most sensitive QOI (counter intuitive) . . . SIAM UQ18 APRIL 16-19, 2018

  24. Example 2: DLR-SynG DLR-SynG dataset ( 159 QOIs, 55 uncertain parameters ) for syngas combustion developed at DLR. Scalar Consistency Vector Consistency • SCM < 0. Analyze ranked sensitivities. – Remove QOI #104 from dataset. SIAM UQ18 APRIL 16-19, 2018

  25. Example 2: DLR-SynG DLR-SynG dataset ( 159 QOIs, 55 uncertain parameters ) for syngas combustion developed at DLR. Scalar Consistency Vector Consistency • SCM < 0. Analyze ranked sensitivities. – Remove QOI #104 from dataset. • SCM < 0. Analyze ranked sensitivities. SIAM UQ18 APRIL 16-19, 2018

  26. Example 2: DLR-SynG DLR-SynG dataset ( 159 QOIs, 55 uncertain parameters ) for syngas combustion developed at DLR. Scalar Consistency Vector Consistency • SCM < 0. Analyze ranked sensitivities. – Remove QOI #104 from dataset. • SCM < 0. Analyze ranked sensitivities. – Remove QOI # 109. SIAM UQ18 APRIL 16-19, 2018

  27. Example 2: DLR-SynG DLR-SynG dataset ( 159 QOIs, 55 uncertain parameters ) for syngas combustion developed at DLR. Scalar Consistency Vector Consistency • SCM < 0. Analyze ranked sensitivities. – Remove QOI #104 from dataset. • SCM < 0. Analyze ranked sensitivities. – Remove QOI # 109. Repeat until consistent SIAM UQ18 APRIL 16-19, 2018

Recommend

Sparse resolutions to inconsistent datasets using L1-minimization Arun Hegde Wenyu Li Jim

Sparse resolutions to inconsistent datasets using L1-minimization Arun Hegde Wenyu Li Jim

Sparse resolutions to inconsistent datasets using L1-minimization Arun Hegde Wenyu Li Jim Oreluk Andrew Packard Michael Frenklach This project is supported by the U.S. Department of Energy, National Nuclear Security Administration, under

862 views • 48 slides
Lecture 5 : Sparse Models Homework 3 discussion (Nima) Sparse Models Lecture - Reading :

Lecture 5 : Sparse Models Homework 3 discussion (Nima) Sparse Models Lecture - Reading :

Lecture 5 : Sparse Models Homework 3 discussion (Nima) Sparse Models Lecture - Reading : Murphy, Chapter 13.1, 13.3, 13.6.1 - Reading : Peter Knee, Chapter 2 Paolo Gabriel (TA) : Neural Brain Control After class - Project groups

571 views • 28 slides
Coupling 3D radiative transfer models with soil vegetation transfer models for sparse vegetation

Coupling 3D radiative transfer models with soil vegetation transfer models for sparse vegetation

Coupling 3D radiative transfer models with soil vegetation transfer models for sparse vegetation and validating with hyperspectral remote sensing and eddy covariance flux data Hideki Kobayashi Japan Agency for Marine-Earth Science and Technology

585 views • 29 slides
PRESENTATION OF THE DRAFT RESOLUTIONS AND TEXT OF THE DRAFT RESOLUTIONS FIRST AND SECOND

PRESENTATION OF THE DRAFT RESOLUTIONS AND TEXT OF THE DRAFT RESOLUTIONS FIRST AND SECOND

MANAGEMENT REPORT AND DRAFT RESOLUTIONS PRESENTATION OF THE DRAFT RESOLUTIONS AND TEXT OF THE DRAFT RESOLUTIONS FIRST AND SECOND RESOLUTIONS Approval of the separate and consolidated fi nancial statements for fi scal year 2017 In the fi rst

352 views • 10 slides
Resolutions from CLAS12: gemc vs data Harut Avakian (JLab) CLAS Collaboration Meeting Nov 14,

Resolutions from CLAS12: gemc vs data Harut Avakian (JLab) CLAS Collaboration Meeting Nov 14,

Resolutions from CLAS12: gemc vs data Harut Avakian (JLab) CLAS Collaboration Meeting Nov 14, 2019 Studies of resolutions and energy loss using gemc Comparing resolutions in gemc and RGA Parameterizing resolutions Input for

310 views • 16 slides
Feature Matching via Sparse Relaxation Models jiangbo@ahu.edu.cn

Feature Matching via Sparse Relaxation Models jiangbo@ahu.edu.cn

Feature Matching via Sparse Relaxation Models jiangbo@ahu.edu.cn 2018-8-8 Content 1 Introduction 2 Problem formulation Related works 3 Sparse models for matching 4 Conclusion and

673 views • 63 slides
Efficient Algorithms for Learning Sparse Models from High Dimensional Data Yoram Singer Google

Efficient Algorithms for Learning Sparse Models from High Dimensional Data Yoram Singer Google

Efficient Algorithms for Learning Sparse Models from High Dimensional Data Yoram Singer Google Research Machine Learning Summer School, UC Santa Cruz, July 18, 2012 1 Sparsity ? THE HIGHER MINIMUM WAGE THAT WAS SIGNED INTO LAW ... WILL BE

855 views • 74 slides
Sparse and Low-Rank Optimization for Dense Wireless Networks Part I: Models Jun Zhang Yuanming

Sparse and Low-Rank Optimization for Dense Wireless Networks Part I: Models Jun Zhang Yuanming

Sparse and Low-Rank Optimization for Dense Wireless Networks Part I: Models Jun Zhang Yuanming Shi HKUST ShanghaiT ech University 1 GLOBECOM 2017 TUTORIAL Outline of Part I Motivations T woVignettes Structured Sparse Models

1.05k views • 60 slides
Capital &amp; Entrepreneurship HOW TO? Jaime L. Amsel, Ph.D. 1 = Strongly disagree, 2 =

Capital &amp; Entrepreneurship HOW TO? Jaime L. Amsel, Ph.D. 1 = Strongly disagree, 2 =

Psychological Capital &amp; Entrepreneurship HOW TO? Jaime L. Amsel, Ph.D. 1 = Strongly disagree, 2 = disagree, 3 = somewhat disagree, 4 = somewhat agree, 5 = agree, 6 = strongly agree ____ 1. I feel confident in representing my work

482 views • 36 slides
Common New Years Resolutions: G Y M New Years Resolutions Good? Bad? Indifferent?

Common New Years Resolutions: G Y M New Years Resolutions Good? Bad? Indifferent?

Common New Years Resolutions: G Y M New Years Resolutions Good? Bad? Indifferent? Reset Do you make resolutions? What are they? Specific Measurable Attainable Relevant Time-bound Common Christian

399 views • 18 slides
Multiscale Sparse Models in Deep Convolutional Networks Tomas Angls, Roberto Leonarduzi,

Multiscale Sparse Models in Deep Convolutional Networks Tomas Angls, Roberto Leonarduzi,

Multiscale Sparse Models in Deep Convolutional Networks Tomas Angls, Roberto Leonarduzi, Stphane Mallat, Louis Thiry, John Zarka, Sixin Zhang Collge de France cole Normale Suprieure Flatiron Institute www.di.ens.fr/data Deep

356 views • 33 slides
Sparse Coding and Dictionary Learning for Image Analysis Part IV: New sparse models Francis

Sparse Coding and Dictionary Learning for Image Analysis Part IV: New sparse models Francis

Sparse Coding and Dictionary Learning for Image Analysis Part IV: New sparse models Francis Bach, Julien Mairal, Jean Ponce and Guillermo Sapiro ICCV09 tutorial, Kyoto, 28th September 2009 Francis Bach, Julien Mairal, Jean Ponce and

353 views • 19 slides
Accelerating Sparse DNN Models without Hardware-Support via Tile-Wise Sparsity 2020/11

Accelerating Sparse DNN Models without Hardware-Support via Tile-Wise Sparsity 2020/11

Accelerating Sparse DNN Models without Hardware-Support via Tile-Wise Sparsity 2020/11 Shanghai Accelerating Sparse DNN Models without Hardware-Support via Tile-Wise Sparsity Cong Guo 1 , Bo Yang Hsueh 2 , Jingwen Leng 1 , Yuxian Qiu 1 ,

420 views • 27 slides
Structured Prediction Problem Unstructured prediction Structured prediction Part of

Structured Prediction Problem Unstructured prediction Structured prediction Part of

Laplace Max-margin Markov Networks Recent Recent A Advances in dvances in L Learning earning SPARSE SPARSE Structured S tructured I I/ /O O M Model odels s: : models, algorithms, and applications models, algorithms, and

629 views • 24 slides
Machine Learning and Sparsity Klaus-Robert Mller !!et al.!! Todays Talk sensing, sparse

Machine Learning and Sparsity Klaus-Robert Mller !!et al.!! Todays Talk sensing, sparse

Machine Learning and Sparsity Klaus-Robert Mller !!et al.!! Todays Talk sensing, sparse models and generalization interpretabilty and sparse methods explaining for nonlinear methods Sparse Models &amp; Generalization? Machine

458 views • 25 slides
Compressed Sensing Meets Machine Learning - Classification of Mixture Subspace Models via Sparse

Compressed Sensing Meets Machine Learning - Classification of Mixture Subspace Models via Sparse

Introduction Classification via Sparse Representation Distributed Pattern Recognition Conclusion Compressed Sensing Meets Machine Learning - Classification of Mixture Subspace Models via Sparse Representation Allen Y. Yang

861 views • 31 slides
Sparse Logistic Regression Learns All Discrete Pairwise Graphical Models Shanshan Wu , Sujay

Sparse Logistic Regression Learns All Discrete Pairwise Graphical Models Shanshan Wu , Sujay

Sparse Logistic Regression Learns All Discrete Pairwise Graphical Models Shanshan Wu , Sujay Sanghavi, Alex Dimakis University of Texas at Austin Graphical models are used to describe complex dependency structures This is a Markov model

395 views • 10 slides
Tutorial: Sparse Recovery Using Sparse Matrices Piotr Indyk MIT Problem Formulation

Tutorial: Sparse Recovery Using Sparse Matrices Piotr Indyk MIT Problem Formulation

Tutorial: Sparse Recovery Using Sparse Matrices Piotr Indyk MIT Problem Formulation (approximation theory, learning Fourier coeffs, linear sketching, finite rate of innovation, compressed sensing...) Setup: Data/signal in n-dimensional

418 views • 25 slides
Sparse Shrunk Additive Models Guodong Liu(University of Pittsburgh) , Hong Chen (Huazhong

Sparse Shrunk Additive Models Guodong Liu(University of Pittsburgh) , Hong Chen (Huazhong

Sparse Shrunk Additive Models Guodong Liu(University of Pittsburgh) , Hong Chen (Huazhong Agricultural Univerisity), Heng Huang (University of Pittsburgh) June 14, 2020 1. Motivation Deep models have made great progress in learning large

315 views • 30 slides
Crafting Quality Resolutions: ALA Midwinter Council Orientation Edition ALA Resolutions Committee

Crafting Quality Resolutions: ALA Midwinter Council Orientation Edition ALA Resolutions Committee

Crafting Quality Resolutions: ALA Midwinter Council Orientation Edition ALA Resolutions Committee Elissia Buell Tyler Dzuba Kathy Hicks-Brooks Ben Allen Hunter David A. Hurley Kimberly Anne Patton Brenda Pruitt-Annisette Marie Pyko Rachel

370 views • 8 slides
Regularization and shrinkage for model selection in sparse GLM models. Challenging problems in

Regularization and shrinkage for model selection in sparse GLM models. Challenging problems in

Regularization and shrinkage for model selection in sparse GLM models. Challenging problems in Statistical Learning Workshop A. Antoniadis LJK-Universit e Joseph Fourier. Grenoble, March 17 &amp; 18, 2011 0-0 Thresholding and

809 views • 76 slides
Sparse Matrices Example Of Sparse Matrices diagonal tridiagonal sparse many elements are

Sparse Matrices Example Of Sparse Matrices diagonal tridiagonal sparse many elements are

Sparse Matrices Example Of Sparse Matrices diagonal tridiagonal sparse many elements are zero lower triangular (?) dense few elements are zero These are structured sparse matrices. May be mapped into a 1D array so that a mapping

391 views • 6 slides
Lecture 9 When The CRC and TCP Checksum Disagree Jonathan Stone, Craig Partridge Advanced

Lecture 9 When The CRC and TCP Checksum Disagree Jonathan Stone, Craig Partridge Advanced

Lecture 9 When The CRC and TCP Checksum Disagree Jonathan Stone, Craig Partridge Advanced Operating Systems 30 November, 2011 SOA/OS Lecture 9, When The CRC and TCP Checksum Disagree 1/26 Introduction Looking for errors Results

309 views • 26 slides
Sparse Matrices sparse many elements are zero dense few elements are zero Example Of

Sparse Matrices sparse many elements are zero dense few elements are zero Example Of

Sparse Matrices sparse many elements are zero dense few elements are zero Example Of Sparse Matrices diagonal tridiagonal lower triangular (?) These are structured sparse matrices. May be mapped into a 1D array so that a mapping

698 views • 11 slides

More recommend