Bayesian Search in Unknown Landscapes, Optimal Stopping and Gittins Indices: chapter answers
Look at about 37 % of the applicants without hiring, then hire the first one better than everyone before, and you choose the best with probability about 37 %: this "37 % rule" is widely repeated, but the conditions under which it holds are demanding. This book uses statistical decision theory to examine three related questions: when to stop observing and act (stopping times), where to look (Bayesian search), and how much immediate cost to accept for the sake of information (the exploration–exploitation trade-off). It covers the exact optimal threshold and success probability of the secretary…
DOI 10.5281/zenodo.23810102 · Read the monograph · 中文版
- Chapter 1 · The Secretary Problem and 1/e: When to Stop If You See Only Ranks and Cannot Go Back
With a known number of applicants, random order, relative ranks only, no recall and only the best counting as success, the optimal rule is a cutoff rule: skip about $n/e$ applicants, then hire the first relative best; t… - Chapter 2 · Variants of the Secretary Problem and How People Stop
When scores are visible and their distribution is known, the optimal threshold changes with the number of applicants remaining and the success probability is about 0.58. With expected rank as the objective, the optimal… - Chapter 3 · Bayesian Search: Where to Look, and What Not Finding Means
Search theory writes "where to look" in terms of three ingredients: the prior, the detection function and Bayesian updating; the greedy rule for the next search picks the place with the largest product of posterior prob… - Chapter 4 · Multi-Armed Bandits: Gittins Indices, Thompson Sampling and the Regret Lower Bound
The Gittins index breaks the multi-armed bandit into one-armed problems: each machine's index is the reward that leaves one indifferent between "continue" and "retire on a fixed reward"; the index exceeds the posterior… - Chapter 5 · Sunk Costs: Rational Persistence, Fallacious Persistence and Stop-Loss Boundaries Set in Advance
The sunk-cost effect is robust in hypothetical scenarios, with a pooled meta-analytic effect size of about 0.5, but in laboratory experiments with real incentives it is small and erratic (surveys and field studies of pr… - Chapter 6 · Applications: Testing Lanes and Selecting People
In lane testing, "earning while learning" and "learning, then earning" are different objectives: Thompson sampling obtains the most conversions during the test, top-two Thompson sampling keeps more trials for challenger… - Chapter 8 · Practice and a Pre-Registered Trial
The sample size rests on the following assumptions: a smallest effect of interest of 4 percentage points (0.04); a standard deviation of each participant's share of 0.12 (the binomial sampling error over 20 games is abo…
Illustrative model calculation, not a live quote