Muutke küpsiste eelistusi

Data-Driven Science and Engineering: Machine Learning, Dynamical Systems, and Control 2nd Revised edition [Kõva köide]

4.49/5 (74 hinnangut Goodreads-ist)
(University of Washington), (University of Washington)
  • Formaat: Hardback, 614 pages, kõrgus x laius x paksus: 259x183x31 mm, kaal: 1390 g, Worked examples or Exercises
  • Ilmumisaeg: 05-May-2022
  • Kirjastus: Cambridge University Press
  • ISBN-10: 1009098489
  • ISBN-13: 9781009098489
Teised raamatud teemal:
Data-Driven Science and Engineering: Machine Learning, Dynamical Systems, and Control 2nd Revised editionSuurem pilt
  • Formaat: Hardback, 614 pages, kõrgus x laius x paksus: 259x183x31 mm, kaal: 1390 g, Worked examples or Exercises
  • Ilmumisaeg: 05-May-2022
  • Kirjastus: Cambridge University Press
  • ISBN-10: 1009098489
  • ISBN-13: 9781009098489
Teised raamatud teemal:
Püsilink: https://www.kriso.ee/db/9781009098489.html
Märksõnad:
Data-driven discovery is revolutionizing how we model, predict, and control complex systems. This text integrates emerging machine learning and data science methods for engineering and science communities. Now with Python and MATLAB®, new chapters on reinforcement learning and physics-informed machine learning, and supplementary videos and code.

Data-driven discovery is revolutionizing how we model, predict, and control complex systems. Now with Python and MATLAB®, this textbook trains mathematical scientists and engineers for the next generation of scientific discovery by offering a broad overview of the growing intersection of data-driven methods, machine learning, applied optimization, and classical fields of engineering mathematics and mathematical physics. With a focus on integrating dynamical systems modeling and control with modern methods in applied machine learning, this text includes methods that were chosen for their relevance, simplicity, and generality. Topics range from introductory to research-level material, making it accessible to advanced undergraduate and beginning graduate students from the engineering and physical sciences. The second edition features new chapters on reinforcement learning and physics-informed machine learning, significant new sections throughout, and chapter exercises. Online supplementary material – including lecture videos per section, homeworks, data, and code in MATLAB®, Python, Julia, and R – available on databookuw.com.

Arvustused

'Finally, a book that introduces data science in a context that will make any mechanical engineer feel comfortable. Data science is the new calculus, and no engineer should graduate without a thorough understanding of the topic.' Hod Lipson, Columbia University 'This book is a must-have for anyone interested in data-driven modeling and simulations. The readers as diverse as undergraduate STEM students and seasoned researchers would find it useful as a guide to this rapidly evolving field. Topics covered by the monograph include dimension reduction, machine learning, and robust control of dynamical systems with uncertain/random inputs. Every chapter contains codes and homework problems, which make this treaties ideal for the classroom setting. The book is supplemented with online lectures, which are not only educational but also entertaining to watch.' Daniel M. Tartakovsky, Stanford University 'Engineering principles will always be based on physics, and the models that underpin engineering will be derived from these physical laws. But in the future models based on relationships in large datasets will be as important and, when used alongside physics-based models, will lead to new insights and designs. Brunton and Kutz will equip students and practitioners with the tools they will need for this exciting future.' Greg Hyslop, Boeing 'Brunton and Kutz's book is fast becoming an indispensable resource for machine learning and data-driven learning in science and engineering. The second edition adds several timely topics in this lively field, including reinforcement learning and physics-informed machine learning. The text balances theoretical foundations and concrete examples with code, making it accessible and practical for students and practitioners alike.' Tim Colonius, California Institute of Technology 'This is a must read for those who are interested in understanding what machine learning can do for dynamical systems! Steve and Nathan have done an excellent job in bringing everyone up to speed to the modern application of machine learning on these complex dynamical systems.' Shirley Ho, Flatiron Institute/New York University

Muu info

A textbook covering data-science and machine learning methods for modelling and control in engineering and science, with Python and MATLAB®.
Preface ix
Acknowledgments xiv
Common Optimization Techniques, Equations, Symbols, and Acronyms xv
Part I Dimensionality Reduction and Transforms
1 Singular Value Decomposition (SVD)
3(50)
1.1 Overview
3(4)
1.2 Matrix Approximation
7(5)
1.3 Mathematical Properties and Manipulations
12(4)
1.4 Pseudo-Inverse, Least-Squares, and Regression
16(7)
1.5 Principal Component Analysis (PCA)
23(5)
1.6 Eigenfaces Example
28(7)
1.7 Truncation and Alignment
35(5)
1.8 Randomized Singular Value Decomposition
40(6)
1.9 Tensor Decompositions and N-Way Data Arrays
46(7)
2 Fourier and Wavelet Transforms
53(44)
2.1 Fourier Series and Fourier Transforms
53(10)
2.2 Discrete Fourier Transform (DFT) and Fast Fourier Transform (FFT)
63(7)
2.3 Transforming Partial Differential Equations
70(6)
2.4 Gabor Transform and the Spectrogram
76(5)
2.5 Laplace Transform
81(4)
2.6 Wavelets and Multi-Resolution Analysis
85(2)
2.7 Two-Dimensional Transforms and Image Processing
87(10)
3 Sparsity and Compressed Sensing
97(34)
3.1 Sparsity and Compression
97(4)
3.2 Compressed Sensing
101(4)
3.3 Compressed Sensing Examples
105(4)
3.4 The Geometry of Compression
109(4)
3.5 Sparse Regression
113(4)
3.6 Sparse Representation
117(3)
3.7 Robust Principal Component Analysis (RPCA)
120(3)
3.8 Sparse Sensor Placement
123(8)
Part II Machine Learning and Data Analysis 131(120)
4 Regression and Model Selection
133(35)
4.1 Classic Curve Fitting
134(6)
4.2 Nonlinear Regression and Gradient Descent
140(5)
4.3 Regression and Ax = b: Over- and Under-Determined Systems
145(6)
4.4 Optimization as the Cornerstone of Regression
151(4)
4.5 The Pareto Front and Lex Parsimoniae
155(3)
4.6 Model Selection: Cross-Validation
158(4)
4.7 Model Selection: Information Criteria
162(6)
5 Clustering and Classification
168(40)
5.1 Feature Selection and Data Mining
169(5)
5.2 Supervised versus Unsupervised Learning
174(4)
5.3 Unsupervised Learning: k-Means Clustering
178(4)
5.4 Unsupervised Hierarchical Clustering: Dendrogram
182(4)
5.5 Mixture Models and the Expectation-Maximization Algorithm
186(3)
5.6 Supervised Learning and Linear Discriminants
189(4)
5.7 Support Vector Machines (SVM)
193(5)
5.8 Classification Trees and Random Forest
198(5)
5.9 Top 10 Algorithms of Data Mining circa 2008 (Before the Deep Learning Revolution)
203(5)
6 Neural Networks and Deep Learning
208(43)
6.1 Neural Networks: Single-Layer Networks
209(5)
6.2 Multi-Layer Networks and Activation Functions
214(5)
6.3 The Backpropagation Algorithm
219(3)
6.4 The Stochastic Gradient Descent Algorithm
222(2)
6.5 Deep Convolutional Neural Networks
224(4)
6.6 Neural Networks for Dynamical Systems
228(5)
6.7 Recurrent Neural Networks
233(3)
6.8 Autoencoders
236(4)
6.9 Generative Adversarial Networks (GANs)
240(2)
6.10 The Diversity of Neural Networks
242(9)
Part III Dynamics and Control 251(136)
7 Data-Driven Dynamical Systems
253(58)
7.1 Overview, Motivations, and Challenges
254(6)
7.2 Dynamic Mode Decomposition (DMD)
260(15)
7.3 Sparse Identification of Nonlinear Dynamics (SINDy)
275(11)
7.4 Koopman Operator Theory
286(10)
7.5 Data-Driven Koopman Analysis
296(15)
8 Linear Control Theory
311(49)
8.1 Closed-Loop Feedback Control
312(5)
8.2 Linear Time-Invariant Systems
317(5)
8.3 Controllability and Observability
322(6)
8.4 Optimal Full-State Control: Linear-Quadratic Regulator (LQR)
328(4)
8.5 Optimal Full-State Estimation: the Kalman Filter
332(3)
8.6 Optimal Sensor-Based Control: Linear-Quadratic Gaussian (LQG)
335(1)
8.7 Case Study: Inverted Pendulum on a Cart
336(10)
8.8 Robust Control and Frequency-Domain Techniques
346(14)
9 Balanced Models for Control
360(27)
9.1 Model Reduction and System Identification
360(1)
9.2 Balanced Model Reduction
361(14)
9.3 System Identification
375(12)
Part IV Advanced Data-Driven Modeling and Control 387(155)
10 Data-Driven Control
389(30)
10.1 Model Predictive Control (MPC)
390(2)
10.2 Nonlinear System Identification for Control
392(6)
10.3 Machine Learning Control
398(10)
10.4 Adaptive Extremum-Seeking Control
408(11)
11 Reinforcement Learning
419(30)
11.1 Overview and Mathematical Formulation
419(7)
11.2 Model-Based Optimization and Control
426(3)
11.3 Model-Free Reinforcement Learning and Q-Learning
429(7)
11.4 Deep Reinforcement Learning
436(4)
11.5 Applications and Environments
440(4)
11.6 Optimal Nonlinear Control
444(5)
12 Reduced-Order Models (ROMs)
449(36)
12.1 Proper Orthogonal Decomposition (POD) for Partial Differential Equations
449(6)
12.2 Optimal Basis Elements: the POD Expansion
455(6)
12.3 POD and Soliton Dynamics
461(4)
12.4 Continuous Formulation of POD
465(5)
12.5 POD with Symmetries: Rotations and Translations
470(5)
12.6 Neural Networks for Time-Stepping with POD
475(4)
12.7 Leveraging DMD and SINDy for Galerkin-POD
479(6)
13 Interpolation for Parametric Reduced-Order Models
485(35)
13.1 Gappy POD
485(5)
13.2 Error and Convergence of Gappy POD
490(3)
13.3 Gappy Measurements: Minimize Condition Number
493(4)
13.4 Gappy Measurements: Maximal Variance
497(3)
13.5 POD and the Discrete Empirical Interpolation Method (DEIM)
500(4)
13.6 DEIM Algorithm Implementation
504(4)
13.7 Decoder Networks for Interpolation
508(4)
13.8 Randomization and Compression for ROMs
512(1)
13.9 Machine Learning ROMs
513(7)
14 Physics-Informed Machine Learning
520(22)
14.1 Mathematical Foundations
520(3)
14.2 SINDy Autoencoder: Coordinates and Dynamics
523(3)
14.3 Koopman Forecasting
526(3)
14.4 Learning Nonlinear Operators
529(4)
14.5 Physics-Informed Neural Networks (PINNs)
533(2)
14.6 Learning Coarse-Graining for PDEs
535(4)
14.7 Deep Learning and Boundary Value Problems
539(3)
Glossary 542(10)
References 552(36)
Index 588
Steven L. Brunton is the James B. Morrison Professor of Mechanical Engineering at the University of Washington and Associate Director of the NSF AI Institute in Dynamic Systems. He is also Adjunct Professor of Applied Mathematics and Computer Science and a Data-Science Fellow at the eScience Institute. His research merges data science and machine learning with dynamical systems and control, with applications in fluid dynamics, biolocomotion, optics, energy systems, and manufacturing. He is an author of three textbooks, and received the UW College of Engineering Teaching award, the Army and Air Force Young Investigator Program (YIP) awards, and the Presidential Early Career Award for Scientists and Engineers (PECASE) award. J. Nathan Kutz is the Robert Bolles and Yasuko Endo Professor of Applied Mathematics at the University of Washington and Director of the NSF AI Institute in Dynamic Systems. He is also Adjunct Professor of Electrical and Computer Engineering, Mechanical Engineering, and Physics and Senior Data-Science Fellow at the eScience Institute. His research interests lie at the intersection of dynamical systems and machine learning. He is an author of three textbooks and has received the Applied Mathematics Boeing Award of Excellence in Teaching and an NSF CAREER award.