•Gradient and Hessian: The analytical gradient 𝜕dB 𝜕𝜆 = 𝜆−𝜇 𝜆(1−𝜆) and Hessian 𝜕2dB 𝜕𝜆2 = 𝜇 𝜆2 + 1−𝜇 (1−𝜆) 2 are provided to the solver for improved convergence

with the following considerations: •Domain constraints: To avoid numerical issues at boundary values where dB(𝜇, 𝜆) → ∞, we constrain 𝜆𝑖,𝑙 ∈ [𝜖,1−𝜖]with𝜖=10 −6 · 2004

1 Pith paper cite this work. Polarity classification is still indexing.

1 Pith paper citing it

browse 1 citing papers

representative citing papers

Selection of the Best Policy under Fairness Constraints for Subpopulations

cs.LG · 2026-05-11 · unverdicted · novelty 7.0

Formalizes SBFC problem and introduces T-a-S-CS algorithm that achieves instance-specific lower bound on sample complexity asymptotically for fair policy selection.

citing papers explorer

Showing 1 of 1 citing paper.

Selection of the Best Policy under Fairness Constraints for Subpopulations cs.LG · 2026-05-11 · unverdicted · none · ref 15
Formalizes SBFC problem and introduces T-a-S-CS algorithm that achieves instance-specific lower bound on sample complexity asymptotically for fair policy selection.

•Gradient and Hessian: The analytical gradient 𝜕dB 𝜕𝜆 = 𝜆−𝜇 𝜆(1−𝜆) and Hessian 𝜕2dB 𝜕𝜆2 = 𝜇 𝜆2 + 1−𝜇 (1−𝜆) 2 are provided to the solver for improved convergence

fields

years

verdicts

representative citing papers

citing papers explorer