Prediction, Learning, and Games

by Nicolo Cesa-Bianchi, Gábor Lugosi

This important text and reference for researchers and students in machine learning, game theory, statistics and information theory offers a comprehensive treatment of the problem of predicting individual sequences. Unlike standard statistical approaches to forecasting, prediction of individual sequences does not impose any probabilistic assumption on the data-generating mechanism. Yet, prediction algorithms can be constructed that work well for all possible sequences, in the sense that their performance is always nearly as good as the best forecasting strategy in a given reference class. The central theme is the model of prediction using expert advice, a general framework within which many related problems can be cast and discussed. Repeated game playing, adaptive data compression, sequential investment in the stock market, sequential pattern analysis, and several other problems are viewed as instances of the experts' framework and analyzed from a common nonstochastic standpoint that often reveals new and intriguing connections.

Genres: Textbooks, Economics, Artificial Intelligence

408 pages, Hardcover

First published March 13, 2006

Reviews

[Prasmit Khanal · Rating 3 out of 5]

I've found that this is the most coherent book for on-line learning. It deals with the subject matter in a very structured form and the book includes a very helpful diagram of a tree which shows shows the chapters relate to one another. Still a very challenging read unless you are used to the writing style of scientific journals.

I found that the maths was hard to follow as assumptions are made based on material discussed in previous chapters but this is not explicitly stated so it made this book hard to keep reading at a steady pace. Maybe its just me and I don't have the level of mathematical skill required to follow along with this book. I understood most things but had to keep flipping back and forth to revisit certain sections.