

Video Lectures: Inference over Networks (25) 
[Books] [
Journals] [
By Topic] [
Conferences]
[Book Chapters][Editorials]
[Theses] [Patents]
[Google Scholar citation page]
[Thomson Reuters Highly Cited Researcher]
Important Copyright Information. Copyright Ali H. Sayed, 2015. All rights reserved. These lectures can be watched on this site but cannot be copied. The lectures can be watched by instructors and students for instructional and educational purposes only. The lectures cannot be downloaded and/or distributed.
Note for Instructors: A solutions manual is available upon request from the author.
 Read
description of course and topics covered.
 Download assignment problems for lecturing.
 Download the eversion of the book. [pdf, 5.7MB]
 Download errata. [pdf]
 Read a description of the book.
Content: These lectures provide a unified and thorough treatment of the theory of distributed adaptation, optimization, and learning by multiagent systems. Necessary background material from linear algebra, matrix theory, and convex function theory is also covered. [more detailed description of course content]
Text Information:
A. H. Sayed, Adaptation, Learning, and Optimization over Networks, Foundations and Trends in Machine Learning, vol. 7, issue 45, NOW Publishers, BostonDelft, 518pp, 2014. ISBN 9781601988508. [pdf, 5.7MB]
Additional Supporting References:
 A. H. Sayed, ``Adaptive networks,'' Proceedings of the IEEE, vol. 102, no. 4, pp. 460497, April 2014. [pdf]
 A. H. Sayed, ``Diffusion adaptation over networks,'' in
Academic Press Library in Signal Processing, vol. 3, R. Chellapa and S. Theodoridis, editors, pp. 323454, Academic Press, Elsevier, 2014. Also available as arXiv:1205.4220 [cs.MA], May 2012. [pdf]
 A. H. Sayed, S.Y. Tu, J. Chen, X. Zhao, and Z. J. Towfic,
``Diffusion strategies for adaptation and learning over networks,'' IEEE Signal Processing Magazine, vol. 30, no. 3, pp. 155171, May 2013. [pdf]
Disclaimer: The lectures have been recorded in an actual class setting with students in attendance. Please excuse imperfections. Lectures 1720 are more demanding than the remaining lectures in the class; they examine the stability and performance of multiagent networks in some technical depth with the necessary proofs and derivations.
 Video Lecture 1 (38 mins): Motivation and Examples.
[slides]
 Video Lecture 2 (26 mins): Complex Gradient Vectors, Appendix A.
[slides]
 Video Lecture 3 (38 mins): Complex Hessian Matrices, Appendix B.
[slides]
 Video Lecture 4 (66 mins): Convex Functions, Appendix C.
[slides]
 Video Lecture 5 (15 mins): Logistic Regression, Appendix G.
[slides]
 Video Lecture 6 (22 mins): MeanValue Theorems, Appendix D.
[slides]
 Video Lecture 7 (60 mins): Lipschitz Conditions, Appendix E.
[slides]

Video Lecture 8 (56 mins): Useful Matrix Results, Appendix F.
[slides]
 Video Lecture 9 (110 mins): Optimization by Single Agents, Chapter 2.
[slides]

Video Lecture 10 (119 mins): Stochastic Optimization by Single Agents, Chapter 3. [slides]
 Video Lecture 11 (98 mins): Stability and LongTerm Dynamics, Secs. 4.14.4. [slides]

Video Lecture 12 (104 mins): Performance by Single Agents, Secs. 4.5 and 4.6. [slides]
 Video Lecture 13 (107 mins): Centralized Adaptation and Learning, Chapter 5. [slides]

Video Lecture 14 (106 mins): MultiAgent Network Model, Chapter 6. [slides]
 Video Lecture 15 (100 mins): MultiAgent Distributed Strategies, Chapter 7. [slides]
 Video Lecture 16 (113 mins): Evolution of MultiAgent Networks, Secs. 8.18.2. [slides]
 Video Lecture 17(110 mins):Stability of MultiAgent Networks,Secs.8.38.4,9.19.2. [slides]
(Excuse imperfections in audio. Consider raising the volume, esp. during first 7 mins.)
 Video Lecture 18 (103 mins): MeanError Network Stability, Sec. 9.3. [slides]
 Video Lecture 19 (102 mins): LongTerm Network Dynamics, Chapter 10. [slides]
 Video Lecture 20 (115mins): MultiAgent Network Performance, I, Secs. 11.111.3 [slides]
 Video Lecture 21 (115mins): MultiAgent Network Performance, II, Secs. 11.311.5 [slides]
 Video Lecture 22 (53 mins): Benefits of Cooperation, Chapter 12. [slides]
 Video Lecture 23 (49 mins): Role of Informed Agents, Chapter 13. [slides]
 Video Lecture 24 (64 mins): Combination Policies, Chapter 14. [slides]

Video Lecture 25 (47 mins): Extensions and Conclusions, Chapter 15. [slides]
Assignments: Examples of homework assignments that can go along with these lectures. The problem numbers below are extracted from
List of Problems (version August 2014):
 Homework 1: Solve problems 2, 5, 6, 8, 11, 15, 16, 18, 24, 30, 35.
 Homework 2: Solve problems 38, 39, 43, 46, 48, 52, 55, 56, 57.
 Homework 3: Solve problems 58, 59, 61, 66, 71, 72, 75, 80, 82, 84.
 Homework 4: Solve problems 86, 88, 93, 96, 98, 100, 102, 105, 107, 114.
 Homework 5: Solve problems 116, 118, 120, 122, 124, 127, 130, 131, 132.



