<ul><li>This work focuses on statistical learning with dependent data and square loss in a specific hypothesis class.</li><li>The objective is to find a sharp noise interaction term, or variance proxy, in learning with dependent data.</li><li>The empirical risk minimizer achieves a rate that depends only on the complexity of the class and second-order statistics, termed as a 'near mixing-free rate'.</li><li>The study combines the concept of a weakly sub-Gaussian class with mixed tail generic chaining to compute optimal rates for various problems.</li></ul>

Sharp Rates in Dependent Learning Theory: Avoiding Sample Size Deflation for the Square Loss

Discover more