cloture
Copyright (c) 2026 Cloture Contributors

This software is licensed under the MIT License (see LICENSE).

Third-Party Code Dependencies
-----------------------------
- Rust workspace (`cloture-core`, `cloture-cli`): ZERO external crate dependencies
  (Rust standard library only; verified in Cargo.lock).
- Python SDK (`cloture`): ZERO runtime package dependencies (Python standard
  library ctypes bindings to libcloture_core).

Mathematical & Literature Attributions
--------------------------------------
All algorithms in `cloture` are original clean-room implementations in Rust.
Their statistical guarantees build on the following published literature:

1. Predictable Plug-In Empirical-Bernstein & Hedged ONS Betting Confidence
   Sequences:
   I. Waudby-Smith & A. Ramdas, "Estimating means of bounded random variables
   by betting," JRSS Series B (2024); X. Fan, I. Grama, Q. Liu, "Exponential
   inequalities for martingales with applications," Electron. J. Probab. (2015).

2. Anytime-Valid Mode & Unbounded Consensus Stopping (ASC-D, CITE, Kraft-IUT,
   Deterministic Budget Lock, and Realized vs. Population Certification):
   Sequential e-process / mixture-martingale mode certification and unit-level
   clustered rollout aggregation frameworks (2025-2026).

3. Correlated-Panel Diagnostics:
   L. Kish (1965) design-effect effective sample size, eigenvalue spectral
   effective rank, stratified Condorcet independence null Monte Carlo,
   Krippendorff's alpha, and leave-one-out panel pruning.

4. Multiple Testing & Active Prediction-Powered Inference:
   R. Wang & A. Ramdas, "False discovery rate control with e-values," JRSS
   Series B (2022); A. N. Angelopoulos, S. Bates, C. Fannjiang, M. I. Jordan,
   T. Zrnic, "Prediction-Powered Inference," Science (2023); A. N. Angelopoulos,
   J. C. Duchi, T. Zrnic, "PPI++: Efficient Prediction-Powered Inference" (2023);
   T. Zrnic & E. J. Candès, "Active Statistical Inference" (ICML 2024).

5. Safe Online Aggregation & Conformal Control:
   I. Wintenberger, "Optimal learning with Bernstein Online Aggregation,"
   Machine Learning (2017); A. N. Angelopoulos et al., "Conformal Risk Control"
   (ICLR 2024); M. Zaffran et al., "Adaptive Conformal Predictions for Time
   Series" (AgACI, ICML 2022).
