minimax criterion

minimax criterion

[′min·ə‚maks krī‚tir·ē·ən]
(statistics)
A concept in game theory and decision theory which requires that losses or expected losses associated with a variable that can be controlled be minimized, and thus maximizes the losses or expected losses associated with the variable that cannot be controlled.
Mentioned in ?
References in periodicals archive ?
We showed that under the minimax criterion, comonotonicity is the least favorable dependence structure among the insurable risks.
Bayesian Model and Minimax Criterion. Bayesian criterion [4, 16] is widely adopted in fusion-based detection system; the objective of Bayesian criterion is to minimize the expected system cost or risk in making decisions, which is denoted by E(M) and formally given by
In this section, we employ a suboptimal detection criterion called minimax criterion [16], which is widely adopted to handle unknown and changeable prior probabilities.
A general approach, combining optimization of the minisum and the minimax criteria with the tunable size of the applicability area for the minimax criterion, was proposed in [20].
The minimax criterion appears as an insurance against the worst case because it aims at minimizing the expected loss in the least favorable case [16].
That is, D-optimality is a minimax criterion. The trade-off in minimizing the maximum prediction error is that the average prediction error is higher.
One intuitively logical rationale for decision in the face of significant risk when probability is either unknown or irrelevant is known as the minimax criterion for decision under risk.
The minimax criterion suggests that we choose the act with the smallest maximum possible loss, or alternatively, with the largest minimum profit.
Stable strategy distributions in a zero-sum game satisfy the minimax criterion.