翻訳と辞書 Words near each other ・ Forward ・ Forward (association football) ・ Forward (Ayla Brown album) ・ Forward (Flame album) ・ Forward (Hoobastank album) ・ Forward (ice hockey) ・ Forward (Sri Lanka) ・ Forward (surname) ・ Forward (The Abyssinians album) ・ Forward (Turn album) ・ Forward air control ・ Forward Air Control Development Unit RAAF ・ Forward air control during the Vietnam War ・ Forward air control operations during the Korean War ・ Forward air control operations during World War II ・ Forward algorithm ・ Forward anonymity ・ Forward Anywhere ・ Forward assist ・ Forward auction ・ Forward Bloc (Socialist) ・ Forward caste ・ Forward chaining ・ Forward Challenge ・ Forward Commitment Procurement ・ Forward Communist Party ・ Forward Communist Party (Anandi Mukherji) ・ Forward Communist Party (Joglekar) ・ Forward compatibility ・ Forward contract Dictionary Lists mini英和辞書 mini和英辞書 Webster 1913 Latin-English FOLDOC Wikipedia English ウィキペディア

翻訳と辞書　辞書検索 [ 開発暫定版 ] スポンサード リンク Forward algorithm ： ウィキペディア英語版 Forward algorithm The forward algorithm, in the context of a hidden Markov model, is used to calculate a 'belief state': the probability of a state at a certain time, given the history of evidence. The process is also known as ''filtering''. The forward algorithm is closely related to, but distinct from, the Viterbi algorithm. For an HMM such as this one: this probability is written as $P(x\_t\; |\; y\_\; )$. Here $x(t)$ is the hidden state which is abbreviated as $x\_t$ and $y\_$ are the observations $1$ to $t$. A belief state can be calculated at each time step, but doing this does not, in a strict sense, produce the most likely state ''sequence'', but rather the most likely state at each time step, given the previous history. ==Algorithm== The goal of the forward algorithm is to compute the joint probability $p(x\_t,y\_)$, where for notational convenience we have abbreviated $x(t)$ as $x\_t$ and $(y(1),\; y(2),\; ...,\; y(t))$ as $y\_$. Computing $p(x\_t,y\_)$ directly would require marginalizing over all possible state sequences $\backslash$, the number of which grows exponentially with $t$. Instead, the forward algorithm takes advantage of the conditional independence rules of the hidden Markov model (HMM) to perform the calculation recursively. To demonstrate the recursion, let ::$\backslash alpha\_t(x\_t)\; =\; p(x\_t,y\_)\; =\; \backslash sum\_,y\_)$. Using the chain rule to expand $p(x\_t,x\_,y\_)$, we can then write ::$\backslash alpha\_t(x\_t)\; =\; \backslash sum\_,y\_)p(x\_t|x\_,y\_)p(x\_,y\_)$. Because $y\_t$ is conditionally independent of everything but $x\_t$, and $x\_t$ is conditionally independent of everything but $x\_$, this simplifies to ::$\backslash alpha\_t(x\_t)\; =\; p(y\_t|x\_t)\backslash sum\_)\backslash alpha\_(x\_)$. Thus, since $p(y\_t|x\_t)$ and $p(x\_t|x\_)$ are given by the model's emission distributions and transition probabilities, one can quickly calculate $\backslash alpha\_t(x\_t)$ from $\backslash alpha\_(x\_)$ and avoid incurring exponential computation time. The forward algorithm is easily modified to account for observations from variants of the hidden Markov model as well, such as the Markov jump linear system. 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「Forward algorithm」の詳細全文を読む スポンサード リンク 翻訳と辞書 : 翻訳のためのインターネットリソース Copyright(C) kotoba.ne.jp 1997-2016. All Rights Reserved.