Analysis Seminar: "From Hoeffdings Tail Inequality to Concentration of Measure"
4:10 pm; Neill 5W
Dr. Haijun Li
In this talk I will discuss "the first inequality" used in analyzing concentration of measure phenomena. I will highlight the elementary proof of this inequality outlined by Wassily Hoeffding in 1963, and then extend Hoeffdings idea to develop a method based on conditioning to derive various well-known concentration inequalities. These inequalities illustrate a remarkable insight given by Michel Talagrand in 1996 on concentration of measure phenomena: "A random variable that depends in a smooth way on many independent random variables (but not too much on any of them) is essentially constant".