Split-kl and PAC-Bayes-split-kl Inequalities for Ternary Random Variables

We present a new concentration of measure inequality for sums of independent bounded random variables, which we name a split-kl inequality. The inequality is particularly well-suited for ternary random variables, which naturally show up in a variety of problems, including analysis of excess losses in classification, analysis of weighted majority votes, and learning with abstention. We demonstrate that for ternary random variables the inequality is simultaneously competitive with the kl inequality, the Empirical Bernstein inequality, and the Unexpected Bernstein inequality, and in certain regimes outperforms all of them. It resolves an open question by Tolstikhin and Seldin [2013] and Mhammedi et al. [2019] on how to match simultaneously the combinatorial power of the kl inequality when the distribution happens to be close to binary and the power of Bersntein inequalities to exploit low variance when the probability mass is concentrated on the middle value. We also derive a PAC-Bayes-split-kl inequality and compare it with the PAC-Bayes-kl, PAC-Bayes-Empirical-Bennett, and PAC-Bayes-Unexpected-Bernstein inequalities in an analysis of excess losses and in an analysis of a weighted majority vote for several UCI datasets. Last, but not least, our study provides the first direct comparison of the Empirical Bernstein and Unexpected Bernstein inequalities and their PAC-Bayes extensions. 1 The Binomial tail bound is slightly tighter, but it does not extend to the PAC-Bayes setting [Langford, 2005] . Our split-kl approach can be directly applied to obtain a "split-Binomial-tail" inequality. 36th Conference on Neural Information Processing Systems (NeurIPS 2022).