TE
TechEcho
Home24h TopNewestBestAskShowJobs
GitHubTwitter
Home

TechEcho

A tech news platform built with Next.js, providing global tech news and discussions.

GitHubTwitter

Home

HomeNewestBestAskShowJobs

Resources

HackerNews APIOriginal HackerNewsNext.js

© 2025 TechEcho. All rights reserved.

Computer scientists prove that a 40-year-old algorithm is optimal

201 pointsby kerckeralmost 10 years ago

13 comments

ruggerialmost 10 years ago
Am I reading this wrong? The title is wrong, right?<p>The paper says that if a strongly sub-quadratic solution exists than the exponential time hypothesis (that SAT cannot be solved in subexponential time) is invalidated.<p>That&#x27;s very interesting, but <i>it&#x27;s not a proof</i> that no strongly sub-quadratic solution exists for SED.<p>Note that exponential time hypothesis <i>is strictly stronger than</i> P!=NP. Even if SAT can&#x27;t be solved in poly time, that doesn&#x27;t mean it can&#x27;t be solved in subexponential time. There are functions that lie between polynomial and exponential...<p>Of course, the paper was careful to explain it, but the media summary...<p>Edit: I was interested to learn about the notion of <i>strongly quadratic</i>; there are O(n^2&#x2F;log n) solutions to SED, but this paper is casting doubt on solutions with time complexity O(n^(2-delta)) for any delta &gt; 0. Another commenter mentioned a method to solve SED like this.
评论 #9700617 未加载
评论 #9703038 未加载
评论 #9701894 未加载
评论 #9701532 未加载
necessityalmost 10 years ago
The Wagner-Fischer (1970) algorithm they mention (AKA Needleman-Wunsch, Lowrance-Wagner, etc) is quadratic in time and space. It might never get any better than that in time - an improved version with linear space complexity was presented by Hirschberg in 75 - but that doesn&#x27;t mean execution times can&#x27;t be improved. Today there are various variations of the same family of algorithms with ever improving execution times. Some adapt the algorithm to massively parallel setups with hundreds of GPUs, others use speculative execution, some use approximations, and so forth.
评论 #9699959 未加载
评论 #9701588 未加载
davmrealmost 10 years ago
Dick Lipton (GATech) has a blog post discussing this paper: <a href="https:&#x2F;&#x2F;rjlipton.wordpress.com&#x2F;2015&#x2F;06&#x2F;01&#x2F;puzzling-evidence&#x2F;" rel="nofollow">https:&#x2F;&#x2F;rjlipton.wordpress.com&#x2F;2015&#x2F;06&#x2F;01&#x2F;puzzling-evidence&#x2F;</a>.
mabboalmost 10 years ago
Short version: computer scientists have shown that if the Edit Distance problem can be solved in sub-quadratic time, then SAT can be solved in sub-exponential time, and therefore P=NP.
评论 #9699248 未加载
sanxiynalmost 10 years ago
The article clearly states this, but the title doesn&#x27;t, so N.B.: the proof is conditional on exponential time hypothesis, that SAT requires exponential time. Of course we don&#x27;t know whether it does.
评论 #9699761 未加载
dfanalmost 10 years ago
Here&#x27;s the paper: <a href="http:&#x2F;&#x2F;arxiv.org&#x2F;abs&#x2F;1412.0348" rel="nofollow">http:&#x2F;&#x2F;arxiv.org&#x2F;abs&#x2F;1412.0348</a>
评论 #9699491 未加载
jbapplealmost 10 years ago
The authors of the paper note (though this MIT News summary omits) that there is a well-known algorithm for edit distance that runs is subquadratic time. In fact, it runs in O(n^2&#x2F;log^2 n) on a word RAM. It uses a method sometimes called &quot;Four Russians&quot; or &quot;shaving a log&quot;.<p>Of course, this does not invalidate the results; I mention it only to dispel the notion that a reader may get from this MIT News summary that there is no subquadratic algorithm likely to be found.<p>It&#x27;s also interesting reading; just Google for &quot;edit distance&quot; and &quot;Four Russians&quot; and you&#x27;ll find many summaries.
dietricheppalmost 10 years ago
&gt; But it also means that computer scientists can stop agonizing about whether they can do better.<p>This is incorrect... because a &quot;better&quot; algorithm might have 99.9% accuracy but be millions of times faster.
评论 #9699261 未加载
评论 #9699426 未加载
评论 #9699563 未加载
评论 #9699765 未加载
judemelanconalmost 10 years ago
The paper mentions an algorithm published in 1980 is O( (n&#x2F;log(n))^2 ), which is better than the Wagner-Fischer algorithm mentioned in the article, even if it&#x27;s not strongly subquadratic.
评论 #9700337 未加载
beefmanalmost 10 years ago
The edit distance between two strings L and S, where |L| &gt;= |S|, is at most |L|. For any number n and string W, a DFA can be built which accepts an input IFF its edit distance to W is &lt; n. Such a DFA can be built and run in O(|W|) time.[1]<p>What&#x27;s the complexity of a &#x27;binary search&#x27; for the edit distance between L and S using this method? The DFA construction complexity grows very fast with respect to n, but we only need log(|L|) DFAs. And the search cut points can be skewed to amortize the growth with respect to n (i.e. first n &lt;&lt; |L|&#x2F;2).<p>[1] see page 63 of <a href="https:&#x2F;&#x2F;scholar.google.com&#x2F;scholar?cluster=9946036749686151606" rel="nofollow">https:&#x2F;&#x2F;scholar.google.com&#x2F;scholar?cluster=99460367496861516...</a>
rbanffyalmost 10 years ago
Well... I guess now it&#x27;s up to the hardware engineers to make it faster... ;-)
zerokkalmost 10 years ago
This is so wrong. Just because it&#x27;s NP-hard (optimization problem, right?) it doesn&#x27;t mean that there aren&#x27;t very-very-very good heuristic algorithms that can make the search significantly faster than exponential in the vast majority of the cases. Take a look at SAT solvers and see how deep the rabbit hole really goes.
评论 #9699361 未加载
评论 #9699796 未加载
评论 #9699635 未加载
anacletoalmost 10 years ago
I love &#x27;proofs&#x27; which rely on an unproven component.