This book provides a comprehensive analysis of the tools and techniques used today for designing and modeling of efficient and robust swarm-intelligence based systems: highly (or fully) decentralized, semi-autonomous, highly-scalable infrastructures in various real-life scenarios. Among others, the book reviews the use of the swarm intelligence paradigm in financial investment, blockchain protocols design, shared transportation systems, communication networks, bioinformatics, and military applications. Theoretical and practical limitations of such systems, as well as trade-offs between the various economic and operational parameters of the systems, are discussed. The book is intended for researchers and engineers in the fields of swarm systems, economics, agriculture, nutrition, and operation research.
Preface
1. Recent Developments in the Theory and Applicability of Swarm Search
Yaniv Altshuler
Overview
Applications
Decentralized Intelligence Architectures and The Swarm Paradigm
Limitations
Swarm Search with Communication
The Convergence of Swarm Intelligence and Modern AI Systems
Ethical Considerations and Societal Impact
Opportunities and Future Research
References
2. Modeling and Prediction of Ride Sharing Utilization Dynamics
Tal Altshuler, Yaniv Altshuler, Rachel Katoshevski and Yoram Shiftan
Introduction
Related Work
Dataset and Methodology
Analyzing the Dynamic Ride-Sharing Network
Summary and Future Work
References
Appendix A: Prediction Results
3. The Dynamics of Autonomous Drone Swarms in Surveillance: Exploring Efficiency, Feasibility, and Complexity
Yaniv Altshuler
Introduction
Related Work
The Dynamic “Cooperative Cleaners” Problem
A Protocol-Agnostic Lower Bound for the Cleaning Time
Impossibility Result
The Cleaning Protocol
Experimental Results vs. Lower Bound
Upper Bound for Static Domains (d =∞)
Experimental Results vs. Upper Bound (Static)
Upper Bound for Expanding Domains
Complexity Analysis
Conclusion
References
4. Defending Large-Scale Critical Infrastructures Using a Swarm of Drones
Yaniv Altshuler
Introduction
Related Work
Patrolling System Optimizing-Problem Definitions
Finding the Optimal Number of Drones
Selecting Which Type of Drones to Use
Case Study I-Theoretical Analysis
Case Study II-Real World Transportation Network Monitoring
Transportation Network Dataset
Conclusions
References
5. Socioeconomic Patterns of Twitter Activity
Jacob Levy Abitbol and Alfredo J. Morales
Introduction
Related Work
Data and Methods
Mobility and Hashtag Space
Topic Analysis
Diversity
Conclusion
References
6. From Microbes to Methane: AI-Based Predictive Modeling of Feed Additive Efficacy in Dairy Cows
Yaniv Altshuler, Tzruya Calvão Chebach and Shalom Cohen
Introduction
Materials and Methods
Field Study Design
Results
Microbial Data Analytics
Discussion
References
Index
Biography
Yaniv Altshuler, a leading expert in Artificial Intelligence, is a researcher at the MIT Media Lab, Cambridge, Massachusetts, USA and a dynamic entrepreneur. Over the past decade he has used his expertise to advise leading financial institutes and government agencies worldwide. Yaniv is the author of 3 books on the topic of AI and has published over 80 scientific papers and filed 18 patents. His research has been covered by Harvard Business Review, Financial Times, Communications of the ACM, IEEE Spectrum and others.
