無限洗牌回原狀
Creado el: 8 de enero de 2025
Creado el: 8 de enero de 2025
Buttoning up misfit pairs
I’m thinking about the Poincaré recurrence theorem, which suggests that in a finite measure-preserving dynamical system, certain states must recur infinitely often, given specific conditions. For a deck of cards, if the process is truly random and measure-preserving, considering all permutations, the probability that it eventually returns to the original arrangement is 1.
Assessing recurrence
I’m figuring out the probability of a deck of cards returning to its original order after shuffles. Since each shuffle is independent, the chance of it never returning is near zero, meaning it will almost surely return.
Comparing concepts
I’m noting that the Poincaré recurrence theorem is distinct from the properties of a Markov chain or random walk. It’s interesting to see the unique characteristics of each approach.
Assessing the scenario
Interpreting the Poincaré recurrence theorem for a deck of cards, it's almost certain to return to the original order after infinite shuffles, though the timescale for this event is extraordinarily vast.
Connecting concepts
A deck of cards, with its finite state and random shuffling, matches a measure-preserving dynamical system. This means that according to the Poincaré recurrence theorem, the deck will eventually return to the original order.
Thought about 龐加萊回歸定理與撲克牌 花的時間為 34 秒
簡短結論
若採用「每次洗牌都等機率產生撲克牌所有排列」的理想模型,那麼從機率論或馬可夫鏈的角度,可知「經過無限次洗牌後,幾乎必然(機率為 1)會回到原本的排序」。這樣的結論雖可以用「龐加萊回歸定理」(Poincaré Recurrence Theorem) 或馬可夫鏈的「遍歷性」(ergodicity) 來理解,不過嚴格說來,龐加萊回歸定理本身是針對「測度守恆」的動力系統所下的結論,而隨機洗牌則通常會用馬可夫鏈遍歷性的概念來得到「最終幾乎必然回到任何狀態(包含原排序)」的結果。
撲克牌的排列總數是有限的
一副 52 張撲克牌(以 A-K 為例,假設還區分花色)所有可能的排列共有 種,這是一個非常巨大的但「有限」的數目。
若每次洗牌都能等機率產生任一種排列
馬可夫鏈遍歷性
以機率論計算
龐加萊回歸定理
通常敘述為:在一個「相空間(phase space)有有限測度且測度守恆」的動力系統中,幾乎所有初始狀態(除了測度為 0 的例外集合)都會在未來的某一時刻再次回到(或非常接近)它的初始位置。
洗牌是否真能用龐加萊回歸定理?