For independent bounded random variables X_i ∈ [a_i, b_i], sums deviate sub-Gaussianly: P(|S_n − E S_n| ≥ nt) ≤ 2·exp(−2n²t²/Σ(b_i−a_i)²). Hoeffding 1963. Backbone of PAC-learning bounds, concentration of empirical means, multi-armed…
For independent bounded random variables X_i ∈ [a_i, b_i], sums deviate sub-Gaussianly: P(|S_n − E S_n| ≥ nt) ≤ 2·exp(−2n²t²/Σ(b_i−a_i)²). Hoeffding 1963. Backbone of PAC-learning bounds, concentration of empirical means, multi-armed…