Tag: topic/bayesian-experimental-design

Sep 23, 2026

Q: Multi-armed and contextual bandits, Bayesian optimisation, sequential Bayesian experimental design / deep adaptive design, dynamic treatment regimes with Q- and A-learning, switchback experiments and RLHF all choose actions from accumulating data. Laid out on one map, what is each optimizing, what is the state, and what feedback does it assume?

Sep 23, 2026

Q: What would continuous learning look like in media measurement, given that media has interaction effects and learning all interactions is costly or needs more cells than available techniques support?

Sep 18, 2026

Q: Chinchilla compute-optimal training, optimal media mix under saturating response (ROAS / mROAS, Dorfman–Steiner), and power analysis / experimental design are all budget-allocation problems under diminishing returns. What transfers between them, and what does not?

Sep 18, 2026

Q: How do the ELBO, the Barber–Agakov posterior bound and the marginal / VNMC bounds on expected information gain, the contrastive PCE and ACE bounds, neural ratio estimation and the forward-KL objective of neural posterior estimation relate? Which direction of KL does each use, is each an upper or lower bound, and what failure does that choice cause?

Sep 18, 2026

Expected Information Gain

Sep 18, 2026

Lindley's Information Measure

Sep 18, 2026

Nested Estimation and Nested Monte Carlo

Sep 18, 2026

Sequential and Adaptive BED

Sep 18, 2026

Adaptive Contrastive Estimation (ACE)

Sep 18, 2026

High-Dimensional Design Applications

Sep 18, 2026

Likelihood-Free ACE and Gradient Estimation

Sep 18, 2026

Prior Contrastive Estimation (PCE)

Sep 18, 2026

From Designs to Policies (Deep Adaptive Design)

Sep 18, 2026

Open Challenges and Future Directions

Sep 18, 2026

Approximate Thompson Sampling and Practical Extensions

Sep 18, 2026

Bernoulli Bandit and Thompson Sampling Algorithm

Sep 18, 2026

Contextual and Linear Bandits

Sep 18, 2026

Multi-Armed Bandits and Thompson Sampling - Overview

Sep 18, 2026

Regret Bounds for Thompson Sampling

Sep 18, 2026

UCB and Greedy Algorithms for Bandits

Sep 18, 2026

Convergence Rates and Estimator Selection

Sep 18, 2026

Implicit Likelihood Estimator

Sep 18, 2026

Variational Marginal Estimator

Sep 18, 2026

Variational NMC Estimator

Sep 18, 2026

Variational Posterior Estimator (Barber-Agakov)

Jul 04, 2026

Q: When should continuous media learning use Bayesian experimental design vs Bayesian optimization vs a bandit?

Jul 04, 2026

Q: How would a geo-holdout experiment be encoded as a design ξ and its EIG computed against an MMM posterior?

Jul 01, 2026

Q: How do adstock/carryover dynamics interact with the timing of sequential media experiments (delayed outcomes)?

Jun 27, 2026

Bayesian Experimental Design - Overview

Jun 27, 2026

Unified SGD BOED - Overview

Jun 27, 2026

Information-Theoretic Design Objectives

Jun 27, 2026

Modern Bayesian Experimental Design - Overview

Jun 27, 2026

Optimization and Gradient Schemes for BED

Jun 27, 2026

The Computational Revolution in EIG Estimation

Jun 27, 2026

Second Brain

Explorer

Tag: topic/bayesian-experimental-design

Q: What would continuous learning look like in media measurement, given that media has interaction effects and learning all interactions is costly or needs more cells than available techniques support?

Q: Chinchilla compute-optimal training, optimal media mix under saturating response (ROAS / mROAS, Dorfman–Steiner), and power analysis / experimental design are all budget-allocation problems under diminishing returns. What transfers between them, and what does not?

Expected Information Gain

Lindley's Information Measure

Nested Estimation and Nested Monte Carlo

Sequential and Adaptive BED

Adaptive Contrastive Estimation (ACE)

High-Dimensional Design Applications

Likelihood-Free ACE and Gradient Estimation

Prior Contrastive Estimation (PCE)

From Designs to Policies (Deep Adaptive Design)

Open Challenges and Future Directions

Approximate Thompson Sampling and Practical Extensions

Bernoulli Bandit and Thompson Sampling Algorithm

Contextual and Linear Bandits

Multi-Armed Bandits and Thompson Sampling - Overview

Regret Bounds for Thompson Sampling

UCB and Greedy Algorithms for Bandits

Convergence Rates and Estimator Selection

Implicit Likelihood Estimator

Variational Marginal Estimator

Variational NMC Estimator

Variational Posterior Estimator (Barber-Agakov)

Q: When should continuous media learning use Bayesian experimental design vs Bayesian optimization vs a bandit?

Q: How would a geo-holdout experiment be encoded as a design ξ and its EIG computed against an MMM posterior?

Q: How do adstock/carryover dynamics interact with the timing of sequential media experiments (delayed outcomes)?

Bayesian Experimental Design - Overview

Unified SGD BOED - Overview

Information-Theoretic Design Objectives

Modern Bayesian Experimental Design - Overview

Optimization and Gradient Schemes for BED

The Computational Revolution in EIG Estimation

Variational BOED - Overview