Complex Adaptive Systems: An Introduction to Computational Models of Social Life (Anglais) Broché – 17 avril 2007
Rentrée scolaire 2017 : découvrez notre boutique de livres, fournitures, cartables, ordinateurs, vêtements ... Voir plus.
- Choisissez parmi 17 000 points de collecte en France
- Les membres du programme Amazon Prime bénéficient de livraison gratuites illimitées
- Trouvez votre point de collecte et ajoutez-le à votre carnet d’adresses
- Sélectionnez cette adresse lors de votre commande
Les clients ayant acheté cet article ont également acheté
Aucun appareil Kindle n'est requis. Téléchargez l'une des applis Kindle gratuites et commencez à lire les livres Kindle sur votre smartphone, tablette ou ordinateur.
Pour obtenir l'appli gratuite, saisissez votre numéro de téléphone mobile.
Détails sur le produit
Si vous vendez ce produit, souhaitez-vous suggérer des mises à jour par l'intermédiaire du support vendeur ?
Commentaires client les plus utiles sur Amazon.com
Part I begins with simple examples of complexity and depicts how emergence can stem from the interaction of multiple agents acting semi-autonomously using simple rules. The theme is developed that individual agents (actors) form complex systems when they are interdependent in some way and these systems can generate complex and unpredictable behaviors without the benefit of a central controller. This leads to a brief but important discussion of some counter-intuitive characteristics of complex systems. For example, "adding noise to the system may actually enhance the ability of a system to find superior outcomes" (p. 30). Several examples make these ideas easy to understand and provide the groundwork for introducing agent-based modeling in Part II.
In Part II, chapter 4 renders the important construct of "emergence" which is the defining characteristic of complex adaptive systems. The authors offer an excellent definition of emergence as "individual, localized behavior [that] aggregates into global behavior that is, in some sense, disconnected from its origins" (p. 44).
Chapter 5 (Part III) begins the detailed discussion of agent-based modeling and computation as a theoretical approach to understanding complex systems. Agent-based models are said to have the capacity to produce "surprising results" (p. 67) because of the interaction of numerous random and non-linear combinations of variables.
Part IV develops ideas about modeling social systems. It primarily covers cellular automata without relying on heavy mathematics. While this is a necessary starting point to introduce some important concepts such as self-organized criticality and power-law phenomena (p. 165), cellular automata is a fairly limited approach to modeling human behavior and the book doesn't go much beyond this type of modeling to explain more sophisticated methods. In addition, most human and organizational behaviors don't follow power laws very closely, so these descriptions are informative but can be misleading. However, the authors correctly emphasize that human behavior is characteristically "fat-tailed" which is contrary to common misconceptions that (average) behavior is primarily Gaussian (normally distributed) in nature.
Chapter 7 introduces an interesting but seemingly arbitrary framework of the Buddhist "Eightfold Way". This appears to be a forced rather than a natural fit to how agents act in organizations and is puzzling for its inclusion. Yet, I may be missing something obvious here. So it would seem to be helpful for the authors to better connect this with the rest of the book (or leave it out entirely). The next section moves immediately to a discussion of modeling forest fires, so at least a summary or transition would be helpful.
Chapter 9 includes some interesting, albeit too brief, discussion of criticality in social systems (p. 177). Only one page is devoted to this topic. In contrast, nine pages were devoted to the "Eight-fold Way". Yet criticality in social systems seems to be the primary reason that one would study complexity in the first place. Hopefully, the authors will consider a revision of this book with some improved organization and a much expanded treatment of criticality.
Overall, the authors introduce and effectively define numerous complexity constructs that apply directly to individuals and organizations. This makes the book relatively unique and valuable, separate from its focus on agent-based modeling. Perhaps the modeling component is less useful in practice because the authors posit that only very simple models can be readily validated and used for most real-life problems. Yet, these core concepts are a necessary starting point for any type of agent based modelling initiative. Consequently, I recommend this book to anyone working in this area.
If you found this review helpful please click "Yes".
There is a particular focus on cellular automata models as a means to highlight adaptive agent principles.
The authors show a lot of discipline by keeping each topic tight so that every chapter is relatively meaty. The use of interesting footnotes provides just the right amount of related colour and helps to remind the reader that the authors have a broad experience in the field (as well as a good sense of humour).
Overall a great survey that I will use as a jumping off point back into the primary literature.
The book is a great textbook. Its flow of topics is in the correct order to taking the reader from the problem of why this approach is needed, through talking openly about the widespread criticism of this approach and tries answering it in a logical and intelligent way. It then continues to explaining what is a model and how to construct one and off to some examples that show other important corner stones of the field. I couldn't ask for a better arrangement of such book. The book is relatively easy to follow and can be used as an undergraduate textbook or for researchers who look for a good introduction to the field.
Some minor problems that I stumbled upon while reading are as follow: (1) chapter 5 is extremely important as it tries to discuss the approach's criticism, however the arguments wasn't always convincing. Specifically, I would like to see some examples of problems X that are given to the neoclassical theorists, and see some discussions on their inability to deal with them and how this approach can cope with them. (2) The research problems that are introduced are very simple (as also stated by the authors themselves), I think that another chapter with two or three examples of real problems would make this book more valuable for the more knowledgeable readers (e.g. some of Epstein works). (3) After doing a lot of reading on that topic I am still amazed to find new terminology to similar ideas I think the field will mature and be more comprehensive to newcomers if the terminology will be standardize.
Overall, this book provides a great introduction to the field, easy to follow, great arrangement of topics. Highly recommended.