Different flavours of CUSUM for change point detection.
-
Updated
Nov 3, 2026 - Python
Different flavours of CUSUM for change point detection.
Open problems in AI safety, stated self-contained for statisticians. A companion to Lionel Levine MAIS.
Poster presented at Wayne State University 2017 Graduate and Postdoc Research Symposium
Monte Carlo audit of repeated testing, alpha spending, power, and stopping time
Final model for the ADIA Lab Structural Break Challenge 2026: 23rd of 729 teams, solo, 66.01 points where guessing scores 50. Code, weights, tests and write-up.
Sequential comparison of probabilistic forecasts and anytime-valid inference in R.
Adversarial game-integrity system that detects rigged casino games and advantage players using sequential hypothesis testing (SPRT/CUSUM) and anomaly detection.
Statistical framework solving the A/B testing "peeking problem" using Sequential Probability Ratio Test (SPRT) with interactive Streamlit dashboard.
Official implementation of On the Tightness and Computational Tractability of Higher-Dimensional Confidence Sequences
Lean 4 library for martingale inequalities, anytime-valid inference, and concentration. Apache 2.0.
A/B testing engine with always-valid sequential inference (mSPRT) — peek continuously without inflating false positives. CUPED variance reduction, SRM detection, guardrails. FastAPI + React.
Adaptive estimation of a threshold probability from noisy coins: a distributional minimax law, with the main bounds formalized in Lean 4 and Mathlib.
A time-travel debugger and statistical bisector for AI agents: snapshot every step, fork K clones from a checkpoint, and binary-search for the step where the success rate collapses.
Pre-registration for forward tests: hash-pinned specs, one-look enforcement, held-out guard, anytime-valid confidence sequences, and a trial ledger.
From-scratch A/B experimentation library: power, fixed-horizon inference, CUPED, sequential testing. Validated by simulation, then run against the Upworthy and ASOS experiment archives, ending in analyst memos with ship/no-ship calls.
Sequential attribution of eavesdropping vs. natural noise in BB84 QKD - identifiability limits, robust change-point detection, and security-preserving operation
Simulation framework measuring how often A/B decision rules are wrong. Daily peeking: 5% -> 21.5% false positive rate.
Prompt versioning and A/B testing that refuses to declare a winner from a peeked p-value: sequential boundaries calibrated for the number of looks you actually take.
An original firewall built on real math — provably consistent rules and a tunable, proven bound on false positives.
Sequential hypothesis testing for continuously-monitored quantum systems. Real-time analysis of measurement signals with adaptive stopping criteria. Published in Quantum 8, 1289 (2024).
To associate your repository with the sequential-analysis topic, visit your repo's landing page and select "manage topics."