作者: Byron K. Williams , James D. Nichols
DOI: 10.1086/284182
关键词: Selection (genetic algorithm) 、 Observable variable 、 Econometrics 、 Variable (computer science) 、 Ecology 、 Observable 、 Unobservable 、 Frame (networking) 、 Random variable 、 Action (philosophy) 、 Biology
摘要: A general approach for obtaining solutions to a class of biological optimization problems is provided. The problem one determining the appropriate time take some action, when action can be taken only once during finite frame. also extended cover number other involving animal choice (e.g., mate selection, habitat selection). Returns (assumed index fitness) are treated as random variables with time-specific distributions, and either observable or unobservable at taken. In case returns, organism assumed base decisions on ancillary variable that associated returns. Optimal policies derived both situations their properties discussed. Various extensions considered, including objective functions based returns than mean, nonmonotonic relationships between returns; possible death bef...