Variance Reduction

Alimama Tech

Dec 28, 2022 · Artificial Intelligence

Sustainable Online Reinforcement Learning for Auto-bidding (SORL)

The Sustainable Online Reinforcement Learning (SORL) framework tackles offline inconsistency in auto‑bidding by iteratively gathering safe online data from real ad systems with a Lipschitz‑based exploration method and training a variance‑suppressed conservative Q‑learning policy, achieving safer, more stable, and higher‑performing bids on Alibaba’s platform.

Variance Reductionauto-biddingoffline inconsistency

0 likes · 18 min read

Sustainable Online Reinforcement Learning for Auto-bidding (SORL)

Bitu Technology

Nov 18, 2022 · Fundamentals

Tubi’s Switchback Experiment Platform: Design, Challenges, and Solutions

The article describes Tubi’s internal experimentation platform, explaining how traditional user‑group A/B tests can suffer from network interference and how Switchback experiments—time‑window based designs—address these issues, detailing their implementation, statistical methods, and the practical challenges overcome.

A/B testingData ScienceSwitchback experiments

0 likes · 12 min read

Tubi’s Switchback Experiment Platform: Design, Challenges, and Solutions

Model Perspective

Nov 9, 2022 · Fundamentals

Why Monte Carlo Converges Slowly: Insights from the Law of Large Numbers and Central Limit Theorem

This article explains how the law of large numbers and the central limit theorem underpin Monte Carlo methods, revealing why their convergence rate is low, how significance and confidence levels are defined, and why variance reduction is crucial for efficient simulations.

Law of Large NumbersProbability TheoryVariance Reduction

0 likes · 5 min read

Why Monte Carlo Converges Slowly: Insights from the Law of Large Numbers and Central Limit Theorem

Model Perspective

Sep 16, 2022 · Fundamentals

Why Adding Non‑Confounding Controls Can Boost Causal Estimates (And When They Hurt)

This article explains how adding covariates that are not confounders can reduce outcome variance and improve causal inference, while controlling for variables that only predict treatment may introduce selection bias and inflate estimation error.

Variance Reductioncausal inferencecontrol variables

0 likes · 21 min read

Why Adding Non‑Confounding Controls Can Boost Causal Estimates (And When They Hurt)

Model Perspective

Sep 8, 2022 · Fundamentals

Why Monte Carlo Converges Slowly: Law of Large Numbers & Central Limit Theorem Explained

This article explains how the law of large numbers and the central limit theorem underpin Monte Carlo methods, illustrating their convergence rate, the role of variance reduction, and the practical steps for applying Monte Carlo to both stochastic and deterministic problems.

Law of Large NumbersProbability TheoryVariance Reduction

0 likes · 4 min read

Why Monte Carlo Converges Slowly: Law of Large Numbers & Central Limit Theorem Explained

DataFunTalk

Jul 29, 2021 · Fundamentals

Offline Sampling in AB Testing: Challenges and Experimental Techniques

The article explains offline sampling for AB testing, detailing why it is needed, the main challenges of limited sample size, population heterogeneity, and non‑random interventions, and presents variance‑reduction, stratified sampling, IPW, and matching methods to address these issues.

AB testingVariance Reductioncausal inference

0 likes · 15 min read

Offline Sampling in AB Testing: Challenges and Experimental Techniques

Alimama Tech

Jul 28, 2021 · Product Management

Offline Sampling in AB Testing: Challenges and Experimental Techniques

Offline sampling in A/B testing assigns experimental units such as users or tags before a trial begins, but suffers from limited sample size, high heterogeneity, and non‑random allocation, which can be mitigated by variance‑reduction methods like CUPED, stratified sampling with inverse‑probability weighting, and matching approaches including propensity‑score matching.

AB testingVariance Reductioncausal inference

0 likes · 15 min read