sprt
Measure A Chess Engine - From Mache (μάχη, battle) - Scripts and CI templates for measuring chess engines
Statistical runtime security for autonomous AI agents. Observe, verify, and enforce agent behavior with deterministic policies, semantic classification, Markov modeling, SPRT, and human approvals.
Differentially private sequential probability ratio tests (Michel, Basu & Kaufmann, AISTATS 2026)
Python package for Wald's sequential probability ratio test