Bernstein inequalities (probability theory)

In probability theory, Bernstein inequalities give bounds on the probability that the sum of random variables deviates from its mean. In the simplest case, let X1, ..., Xn be independent Bernoulli random variables taking values +1 and −1 with probability 1/2 (this distribution is also known as the Rademacher distribution), then for every positive , Bernstein inequalities were proven and published by Sergei Bernstein in the 1920s and 1930s. Later, these inequalities were rediscovered several times in various forms. Thus, special cases of the Bernstein inequalities are also known as the Chernoff bound, Hoeffding's inequality and Azuma's inequality.

1. Let be independent zero-mean random variables. Suppose that almost surely, for all Then, for all positive ,
2. Let be independent zero-mean random variables. Suppose that for some positive real and every integer , Then
3. Let be independent zero-mean random variables. Suppose that for all integer Denote Then,
4. Bernstein also proved generalizations of the inequalities above to weakly dependent random variables. For example, inequality (2) can be extended as follows. Let be possibly non-independent random variables. Suppose that for all integers , Then More general results for martingales can be found in Fan et al. (2015). The proofs are based on an application of Markov's inequality to the random variable for a suitable choice of the parameter . The Bernstein inequality can be generalized to Gaussian random matrices. Let be a scalar where is a complex Hermitian matrix and is complex vector of size . The vector is a Gaussian vector of size . Then for any , we have where is the vectorization operation and where is the largest eigenvalue of . The proof is detailed here. Another similar inequality is formulated as where .