TE
科技回声
首页24小时热榜最新最佳问答展示工作
GitHubTwitter
首页

科技回声

基于 Next.js 构建的科技新闻平台,提供全球科技新闻和讨论内容。

GitHubTwitter

首页

首页最新最佳问答展示工作

资源链接

HackerNews API原版 HackerNewsNext.js

© 2025 科技回声. 版权所有。

The Viterbi Algorithm at 50 (2017)

112 点作者 akhayam大约 2 年前

9 条评论

sizzzzlerz大约 2 年前
After graduating with my BSEE and getting my first professional job, one of my very first tasks was to implement this algorithm on a PDP 11, in assembly. I first implemented it in Fortran (C had just come out and I didn't know it yet) to run on a VAX 11/780. Once it was working, it was amazing to see how powerful it was in correcting noisy data, even using rate 1/2 coding and short constraint lengths. Consider that the Voyager spacecraft, now nearly 24 billion miles from earth, uses it and it is still possible to decode it's data. In my mind, Viterbi's algorithm is equal in importance in communications with the FFT.
评论 #35938574 未加载
评论 #35940855 未加载
akhayam大约 2 年前
What is often overlooked is the philanthropic and entrepreneurial impact that Andrew Viterbi had in this world.<p>In 2004, he donated $52 million to University of Southern California (USC) School of Engineering:<p><a href="https:&#x2F;&#x2F;en.wikipedia.org&#x2F;wiki&#x2F;USC_Viterbi_School_of_Engineering" rel="nofollow">https:&#x2F;&#x2F;en.wikipedia.org&#x2F;wiki&#x2F;USC_Viterbi_School_of_Engineer...</a><p>He was also the co-founder of Qualcomm with a market cap of ~$115B today:<p><a href="https:&#x2F;&#x2F;www.google.com&#x2F;finance&#x2F;quote&#x2F;QCOM:NASDAQ" rel="nofollow">https:&#x2F;&#x2F;www.google.com&#x2F;finance&#x2F;quote&#x2F;QCOM:NASDAQ</a><p><a href="https:&#x2F;&#x2F;en.wikipedia.org&#x2F;wiki&#x2F;Qualcomm" rel="nofollow">https:&#x2F;&#x2F;en.wikipedia.org&#x2F;wiki&#x2F;Qualcomm</a>
评论 #35941774 未加载
jchook大约 2 年前
The Viterbi Algorithm can be used to split multi-word #hashtags or any wordssmushedtogetherlikethis with surprising efficiency and accuracy.<p><a href="https:&#x2F;&#x2F;stackoverflow.com&#x2F;a&#x2F;481773" rel="nofollow">https:&#x2F;&#x2F;stackoverflow.com&#x2F;a&#x2F;481773</a><p><a href="https:&#x2F;&#x2F;github.com&#x2F;jchook&#x2F;wordseg">https:&#x2F;&#x2F;github.com&#x2F;jchook&#x2F;wordseg</a>
jwilk大约 2 年前
<a href="https:&#x2F;&#x2F;en.wikipedia.org&#x2F;wiki&#x2F;Viterbi_algorithm" rel="nofollow">https:&#x2F;&#x2F;en.wikipedia.org&#x2F;wiki&#x2F;Viterbi_algorithm</a>
评论 #35938017 未加载
mikewarot大约 2 年前
So, today I learned this algorithm with the help of some youtube videos.<p>It seems to me the primary advantage is to avoid the combinatorial explosion of having to consider the hamming distance of all possible original messages while still getting the correct answer.<p>If you combined this technique with Feynman diagrams which always have 3 paths to a vertex, maybe there are ways to use this to solve quantum situations without the combinatorial explosion?!?!?
评论 #35944504 未加载
supriyo-biswas大约 2 年前
This came up recently for a hobby project of mine where I had to tokenize Japanese sentences for processing. Ultimately, I implemented a workaround, because I don&#x27;t have a signal processing background and implementing the algorithm by understanding the theory was just too complicated.
nullc大约 2 年前
Checkout the film Mr. Nobody (2009) for a (unintentional?) movie in the structure of the Viterbi algorithm.
akhayam大约 2 年前
Seems like a lot of folks have tried starting from the theory and getting stuck in their tracks.<p>Would anyone be interested in building a low-code implementation of Viterbi algorithm? I am happy to collaborate on it as long as we commit to releasing it in open-source.
fastaguy88大约 2 年前
From the &quot;horses&quot; mouth -- Viterbi describing the Viterbi algorithm, and related developments:<p><a href="http:&#x2F;&#x2F;www.youtube.com&#x2F;watch?v=UgW4p7-jpUs">http:&#x2F;&#x2F;www.youtube.com&#x2F;watch?v=UgW4p7-jpUs</a>