Английская Википедия:Bayesian regret

Материал из Онлайн справочника
Перейти к навигацииПерейти к поиску

Шаблон:More footnotes needed In stochastic game theory, Bayesian regret is the expected difference ("regret") between the utility of a Bayesian strategy and that of the optimal strategy (the one with the highest expected payoff).

The term Bayesian refers to Thomas Bayes (1702–1761), who proved a special case of what is now called Bayes' theorem, who provided the first mathematical treatment of a non-trivial problem of statistical data analysis using what is now known as Bayesian inference.

Economics

This term has been used to compare a random buy-and-hold strategy to professional traders' records. This same concept has received numerous different names, as the New York Times notes:

"In 1957, for example, a statistician named James Hanna called his theorem Bayesian Regret. He had been preceded by David Blackwell, also a statistician, who called his theorem Controlled Random Walks.[1] Other, later papers had titles like 'On Pseudo Games',[2] 'How to Play an Unknown Game'[3]Шаблон:Citation needed, 'Universal Coding'[4] and 'Universal Portfolios'".[5][6]

Social Choice (voting methods)

"Bayesian Regret" has also been used as an alternate term for social utility efficiency, that is, a measure of the expected utility of different voting methods under a given probabilistic model of voter utilities and strategies. In this case, the relation to Bayes is unclear, as there is no conditioning or posterior distribution involved.

References

Шаблон:Citation style Шаблон:Reflist

  1. Controlled random walks, D Blackwell, Proceedings of the International Congress of Mathematicians 3, 336-338
  2. Шаблон:Cite journal
  3. Шаблон:Citation
  4. Шаблон:Cite journal
  5. Шаблон:Cite journal
  6. Шаблон:Cite news