Ergodic Theory and Fractal Geometry

Fractals are beautiful and complex geometric objects. Their study, pioneered by Benoit Mandelbrot, is of interest in mathematics, physics and computer science. Their inherent structure, based on their self-similarity, makes the study of their geometry amenable to dynamical approaches. In this book, a theory along these lines is developed by Hillel Furstenberg, one of the foremost experts in ergodic theory, leading to deep results connecting fractal geometry, multiple recurrence, and Ramsey theory. In particular, the notions of fractal dimension and self-similarity are interpreted in terms of ergodic averages and periodicity of classical dynamics; moreover, the methods have deep implications in combinatorics. The exposition is well-structured and clearly written, suitable for graduate students as well as for young researchers with basic familiarity in analysis and probability theory. -Endre Szemeredi, Renyi Institute of Mathematics, Budapest Fractal geometry represents a radical departure from classical geometry, which focuses on smooth objects that 'straighten out' under magnification. Fractals, which take their name from the shape of fractured objects, can be characterized as retaining their lack of smoothness under magnification. The properties of fractals come to light under repeated magnification, which we refer to informally as 'zooming in'. This zooming-in process has its parallels in dynamics, and the varying 'scenery' corresponds to the evolution of dynamical variables. The present monograph focuses on applications of one branch of dynamics-ergodic theory-to the geometry of fractals. Much attention is given to the all-important notion of fractal dimension, which is shown to be intimately related to the study of ergodic averages. It has been long known that dynamical systems serve as a rich source of fractal examples. The primary goal in this monograph is to demonstrate how the minute structure of fractals is unfolded when seen in the light of related dynamics.