Muutke küpsiste eelistusi

AI for Physics [Pehme köide]

  • Formaat: Paperback / softback, 129 pages, kõrgus x laius: 198x129 mm, kaal: 149 g, 7 Line drawings, black and white; 3 Halftones, black and white; 10 Illustrations, black and white
  • Sari: AI for Everything
  • Ilmumisaeg: 29-Aug-2022
  • Kirjastus: CRC Press
  • ISBN-10: 1032151692
  • ISBN-13: 9781032151694
Teised raamatud teemal:
  • Formaat: Paperback / softback, 129 pages, kõrgus x laius: 198x129 mm, kaal: 149 g, 7 Line drawings, black and white; 3 Halftones, black and white; 10 Illustrations, black and white
  • Sari: AI for Everything
  • Ilmumisaeg: 29-Aug-2022
  • Kirjastus: CRC Press
  • ISBN-10: 1032151692
  • ISBN-13: 9781032151694
Teised raamatud teemal:
Written in accessible language without mathematical formulas, this short book provides an overview of the wide and varied applications of artificial intelligence (AI) across the spectrum of physical sciences. Focusing in particular on AI's ability to extract patterns from data, known as machine learning (ML), the book includes a chapter on important machine learning algorithms and their respective applications in physics. It then explores the use of ML across a number of important sub-fields in more detail, ranging from particle, molecular and condensed matter physics, to astrophysics, cosmology and the theory of everything. The book covers such applications as the search for new particles and the detection of gravitational waves from the merging of black holes, and concludes by discussing what the future may hold.

Written in accessible language without mathematical formulas, the book covers important machine learning algorithms and their respective applications in physics and across a number of important sub-fields in more detail, ranging from particle, molecular and condensed matter physics, to astrophysics, cosmology and the theory of everything.

Arvustused

"AI for Physics is a very recommendable, easy-to-read and wide-ranging review of applications that artificial intelligence can have in many of the branches of physics. Across its pages, Dr. Knecht, assisted by nine experts in the different fields, goes over the wide spectrum of scales in which machine learning can contribute to the research in physics, from the subatomic world to the whole universe as in cosmology. The book is written in very accessible language, which makes it a very good text for beginners that want to start learning about these subjects, providing them a general view of the current state of this very promising cooperation between AI and physics. But it is also a very interesting reading for those who, having certain knowledge in some of the applications of AI in physics, want to complete their view of the whole picture, and benefit from the big amount of good references spread along its lines, that make it easy to follow the reading with more specific texts."

-Dr. Óscar de Abril, Associate Professor, Technical University of Madrid, Spain

"Artificial intelligence is on a rapid path of revolutionizing science, upsetting long-established routines, and learning complex relationships in nature beyond what human minds can comprehend. AI for Physics hits the nail on the head in explaining how that is possible and where the future may take us. A great read for scientists and anybody else curious about how soon computers will be smarter than us."

-Michael Feig, Professor, Michigan State University

"In an "afternoon read", this new book, aimed at non-specialists, intends to update its readers on the emerging applications of machine learning in physics. Without equations and derivations, the book provides an intuitive understanding of what seems to be going on, while simultaneously reviewing many of the exciting applications of ML in physics, ranging from the astronomically large to the femto-meter sized small. A wealth of references allows to all these topics further self- study. I can readily recommend the book to all scientists interested in how ML is already shaping the way of science. Graduate students will find it especially interesting to quickly get an overview if ML has already entered their chosen research topic."

- Rudolf A. Roemer, Professor, University of Warwick

"Very well written, I have enjoyed reading it! "

- Parimal Kar, Associate Professor, Indian Institute of Technology Indore.

Acknowledgments xi
Contributors xiii
List of Abbreviations
xvii
Part I Opening
1 Gathering the Team
3(8)
Volker Knecht
AI and Machine Learning
3(2)
A Brief History of Physics
5(3)
References
8(3)
2 Teamplay
11(16)
Volker Knecht
Machine Learning Physics
11(3)
Impact of Physics on Machine Learning
14(1)
Statistical Physics of ML
15(1)
Analog Computers
16(1)
Quantum Computers
17(2)
Machine Learning the Physical World from Subatomic to Cosmic Scales
19(4)
References
23(4)
3 The Rules of the Game
27(18)
Volker Knecht
Kilian Hikaru Scheutwinkel
Supervised Learning
28(1)
Classification versus Regression
28(1)
Simple Mappings
29(1)
Complex Mappings
29(2)
River Deep -- Mountain High
31(1)
Choosing the Number of Parameters as a Balancing Act
32(1)
Bias-Variance Trade-off
33(1)
Kernel Methods
33(1)
Decision Trees
34(1)
Artificial Neural Networks
34(2)
Treating Uncertainty and Prior Knowledge: Bayesian Inference
36(1)
Symbolic Regression
37(1)
Unsupervised Learning
37(1)
Clustering and Principal Component Analysis
38(1)
Autoencoders
38(1)
Physics-Inspired Algorithm: Restricted Boltzmann Machine
39(1)
Generative Adversarial Networks
39(1)
Reinforcement Learning
40(1)
What's Next?
40(1)
References
40(5)
Part II Machine-Learning the World from Subatomic to Cosmic Scales
4 AI for Particle Physics
45(14)
Mario Campaneixi
Volker Knecht
The Standard Model
46(2)
Open Problems
48(1)
Theories beyond SM
49(1)
Machine Learning Particle Physics
49(1)
Cut-Based Event Selection in a Particle Physics Experiment
50(1)
Particle and Event Selection with Neural Networks and Boosted Decision Trees
50(1)
Machine Learning for Jet Physics
51(3)
Convolutional Neural Networks for Neutrino Experiments
54(2)
References
56(3)
5 AI for Molecular Physics
59(12)
Mayank Agrawal
Volker Knecht
Speeding Up Simulations I: Machine Learning Atomistic Force Fields
61(2)
Using Machine Learning to Analyze Output of Simulations
63(3)
Speeding Up Simulations II: Machine Learning Coarse-Grained Force Fields
66(2)
References
68(3)
6 AI for Condensed Matter Physics
71(12)
Alvaro Diaz Hernandez
Chao Fang
Volker Knecht
Using Machine Learning to Overcome Sampling Problem for Spin Glasses
72(3)
Machine Learning Topological Order Transition
75(2)
Machine Learning Quantum Many-Body Systems
77(1)
Looking from Outside: Machine Learning Quantum Tomography
78(1)
Machine Learning Based Design of New Materials and Quantum States
78(2)
References
80(3)
7 AI for Cosmology
83(20)
Kilian Hikaru Scheutwinkel
Daniel Grun
Bernard Jones
Jimena Gonzalez Lozano
Volker Knecht
The Concordance Model of Cosmology
83(2)
Machine Learning Big Data and the Global Shape of the Universe
85(2)
Machine Learning New Physics versus Instrumental Effects
87(1)
Machine Learning Photometric Redshift
88(1)
Objects in the Mirror May Be Bluer than They Appear
88(2)
AI to the Rescue - But with the Right Architecture and Training
90(1)
Machine Learning Cosmic Structure
91(1)
Bubble Universes All the Way Down
91(1)
Distortion Probes Gravitation: Interstellar Lensing
92(1)
Fishing for Complements with the Cosmic Web
93(1)
Machine Learning Gravitational Waves
94(4)
Note
98(1)
References
98(5)
Part III Showdown
8 AI for Theory of Everything
103(12)
Yang-Hui He
Volker Knecht
Physics and Geometry
103(1)
String Theory
104(1)
Extra Dimensions
104(1)
Why String Theory?
105(1)
Machine-Learning the Landscape
106(1)
The String Landscape and Vacuum Degeneracy Problem
107(2)
More on Machine-Learning the Landscape
109(4)
Epilogue
113(1)
References
113(2)
9 Conclusion and Outlook
115(2)
Volker Knecht
References 117(2)
Appendix: Table of Contents for Electronic Supplement 119(2)
Index 121
Volker Knecht, Germany, Editor at International Journal of Molecular Sciences, Science Writer as Freelancer. Diploma in Physics at University of Kaiserslautern, PhD in Theoretical Physics at University of Göttingen, PhD project at MPI Göttingen, postdoc at University of Groningen, group leader and PI at MPI Potsdam and University of Freiburg. Research at the interface between physics, chemistry, biology, and computer science for 17 years.