|
Dyadic Encoding is a form of binary encoding defined by Smullyan commonly used in computational complexity theory '1's and '2's that is bijective and has the "technical advantage, not shared by binary, of setting up a one-to-one correspondence between finite strings and numbers."〔(Classes of Predictable Computable Functions by Robert W. Ritchie )〕 Dyadic encoding works by using a recursive definition of concatenating strings of '1's and '2's together using the following formula. * dya(0) = ξ (empty set) * dya(2n + 1) = dya(n)'1' ''Odd numbers'' * dya(2n + 2) = dya(n)'2' ''Even numbers'' For example: == References == Computational complexity theory 抄文引用元・出典: フリー百科事典『 ウィキペディア(Wikipedia)』 ■ウィキペディアで「Dyadic Encoding」の詳細全文を読む スポンサード リンク
|