方法对比

	自适应A/B测试	AB设计
领域	实验设计	实验设计
方法族	Process / pipeline	Process / pipeline
起源年份	1952 (Robbins); applied to A/B testing from ~2010s onward	1960s
提出者	Herbert Robbins (bandit framework); Thompson Sampling formalized by William R. Thompson	Murray Sidman; Baer, Wolf & Risley
类型	Adaptive experimental design	Single-subject experimental design
开创性文献	Russo, D., Van Roy, B., Kazerouni, A., Osband, I., & Wen, Z. (2018). A Tutorial on Thompson Sampling. Foundations and Trends in Machine Learning, 11(1), 1–96.	Sidman, M. (1960). Tactics of Scientific Research: Evaluating Experimental Data in Psychology. Basic Books.
别名	adaptive AB test, bandit A/B test, multi-armed bandit testing, online adaptive experiment	baseline-intervention design, AB single-case design, AB phase design
相关	6	4
摘要	An Adaptive A/B test is an experimental design that dynamically reallocates traffic or participants toward better-performing variants during the experiment itself, rather than holding allocations fixed until the end. Drawing on multi-armed bandit algorithms such as Thompson Sampling or Upper Confidence Bound (UCB), it balances the exploration of uncertain variants with the exploitation of those already showing superior performance, typically yielding higher aggregate outcomes while still producing valid inferential conclusions.	The AB design is the simplest single-subject experimental design, consisting of two sequential phases: a baseline phase (A) in which the target behavior is observed under natural conditions without intervention, followed by an intervention phase (B) in which the treatment or manipulation is introduced. Changes in the behavior's level, trend, or variability between phases are used to infer the effect of the intervention on the individual participant.