Simple Heuristics That Make Us Smart (Evolution and Cognition) [Englisch] [Taschenbuch]

As a note, i'm picky when it comes both to writing and thinking. And i hate most books written by academics. Even the ones with good information (eg, Fodor's Modularity) are hard to read and filled with confusing, field-specific words. Not this book. It's really well written. Written in plain English, very few assumptions, very thorough analysis, lots of self-criticism, lots and lots of data (OK, that part is boring and can be skipped, but it's comforting to know it's there)

What's it about? Common AI, psych and economic decision and learning algorithms (decision trees, neural nets, Bayes, multiple linear regression, etc.) are compared to several absurdly simple algorithms the authors believe real humans use. The various approaches are compared and evaluated on the basis of performance, accuracy on training data, accuracy on test data (generalization) and amount of input data required. Tests are on the standard UC Irvine data learning test sets. Comparisions, outcome explanations and relevance to the human mind and the real world are provided. Explanations and analysises are easy to understand and pretty convincing

i've decided to use a lot of what was in this book in my software, things that have made my agents more natural and easier to implement. i absolutely love this book

The ensuing pages compare several theoretical models, such as Multiple Linear Regression and Dawes Rule to their own Take the First and Take the Best models.

Most of the tests were simulated on a computer. You would feed each decision making model into the computer, and then feed in various data for it to make decisions on. One popular test is "Which is the most populated German City." The computer had data on various German Cities with populations over 100,000. It also had several indicators, such as whether it has a soccer team, or a rail system, or is a state capital. The system would present two cities, with the indicators, and the decision making model would figure out which was the most populous one.

Right now I'm in a chapter called "Bayesian Benchmarks for Fast and Frugal Heuristics." It's about halfway through the book, and I'm not sure I'll finish. While the second half sounds interesting, this book is highly academic and the authors are concerned with presenting proofs for everything they say, in detail. Sort of like a victorian novel that starts of by telling you what it's going to tell you, and then tells you several times. I may skim it because I do find the subject matter intereting.

I certainly don't regret buying this book, having mathematical models for decision making is certainly handy (as someone interested in AI), but I wouldn't call it light reading, nor would I reccomend it to a manager interested in the decision making process.

I found much more interesting "Sources of Power" by Gary Klein. Indeed, I consider Sources of Power to be one of the most informative and most entertaining books I've ever read, and wish more like it existed.

In summation, I found this book to be highly academic and theoretical. If you are a human being interested in the decision making process as it is carried out by humans, I reccomend the more hands-on Sources of Power by Gary Klein. If you are interested in simple, statistical models for decision making (the kind you can teach a computer), then pick up this book.

The 18 authors from various academic fields believe that decision rules and the environment in which they are used should always be considered together. Moreover it seems plausible that a simple rule which performs as well as a rule that requires more effort to apply, should be the preferred way of explaining the observed behavior.

The authors propose a bunch of simple heuristics for all kinds of problems. One particularly impressive example was the extremely simple "recognition heuristic" which e.g. proved to be quite successful on the stock market. For all heuristics in the book it is shown that they are easy to use, that they require little memory and computational capacity, and that therefore they appear to be very plausible models for explaining human (and animal) behavior.

If you are interested in decision making and/or if you are working in the fields of psychology, economics, artificial intelligence or related fields, this book is a "must-have"!