next up previous
次へ: Detection of the number 上へ: Multi-pitch Detection Algorithm 戻る: Multi-pitch Detection Algorithm

Criterion of Model Selection

Provided multiple different model candidates exist, the optimal model must somehow be judged. Here we introduce Akaike Information Criterion (AIC) which was proposed by Akaike in 1973 [9]. AIC is given by

$\displaystyle \textmc{AIC}\!=\!-2\!\times\!(\textmc{maximum log-likelihood of model})$      

$\displaystyle \hspace{8ex}+2\!\times\!(\textmc{number of free parameters of model}),$     (9)

whose minimum offers a proper estimate of the number of free parameters.