Beta Distribution Parameter Updates — DEJAN Codebase Explorer

After each round, the tested hypothesis receives a binary reward where \(\alpha\) increases by one if the edit improves the target's rank over the current best, while \(\beta\) increases by one if the rank remains unchanged or worsens.

Referenced by

Thompson Sampling in Content Optimizer
Claim Status Evolution Rules
Thompson Sampling vs UCB Formulation
Rank Change Evaluation
Beta Distributions in Content Optimizer
Stage 1 Diversity Selection
Bayesian Posterior Update Mechanics
Beta Distribution Posterior
Claim Elicitation and Seeding
Two-Stage Bayesian Acquisition
Claim Status Transitions
Freshness-Based Diversity Weight
Content Optimizer Acquisition Algorithm
Two-Stage Acquisition Policy
Content Optimizer Four-Stage Pipeline
Thompson Sampling Selection Mechanics
Thompson Sampling Optimization
Beta distribution claim updates
Content Optimizer Acquisition Algorithm
Optimizer Testable Hypotheses
Multi-Round Optimization Cycle
Rank Factor Classes