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。

When does a physical system compute?

93 点作者 yiransheng大约 11 年前

9 条评论

j2kun大约 11 年前
I don&#x27;t see what this paper is contributing other than the assertion that a physical system needs to functorially agree with the mathematical model that represents it. And that if that model is Turing complete, or represents some useful subset of computational functionality, then the physical system &quot;computes.&quot; The fact that a sitting rock does nothing but sit is a perfectly fine, albeit severely limited, computational system (it even satisfies their definition of a computing physical system!). All of this just seems...obvious.<p>It looks like their definitions are so general that they don&#x27;t answer their own questions, which by their main motivations (section XI) are fundamentally problems of construction and scale. It&#x27;s not enough that there exists a theory that proves your physical system computes, you have to know the representation explicitly and be able to scale the system arbitrarily. [Edit:] The problem being that this doesn&#x27;t show up in their actual definitions.<p>Maybe I&#x27;m reading it too shallowly, though. Can someone who&#x27;s more well-versed in the background of this paper explain it better? It also doesn&#x27;t help my confidence that they seem to completely ignore the rest of computer science (mentioning Turing machines only as an aside, and as part of a false assertion about (quantum) Turing machines being the only universal logical systems).
评论 #7696004 未加载
评论 #7696234 未加载
wfn大约 11 年前
Two somewhat-related-pointers for people who decide that they&#x27;re curious about these things:<p>* a light and nice read: <a href="https://www.frc.ri.cmu.edu/~hpm/project.archive/general.articles/1998/SimConEx.98.html" rel="nofollow">https:&#x2F;&#x2F;www.frc.ri.cmu.edu&#x2F;~hpm&#x2F;project.archive&#x2F;general.arti...</a> (an oft-cited article)<p>* the madness of Max Tegmark: <a href="http://arxiv.org/abs/grqc/9704009" rel="nofollow">http:&#x2F;&#x2F;arxiv.org&#x2F;abs&#x2F;grqc&#x2F;9704009</a> , <a href="http://arxiv.org/abs/0704.0646" rel="nofollow">http:&#x2F;&#x2F;arxiv.org&#x2F;abs&#x2F;0704.0646</a> (cliff&#x27;s notes: &quot;The [...] postulate in this theory is that all structures that exist mathematically exist also physically, by which we mean that in those complex enough to contain self-aware substructures (SASs), these SASs will subjectively perceive themselves as existing in a physically ``real&#x27;&#x27; world.&quot; Which means: take <a href="http://xkcd.com/505/" rel="nofollow">http:&#x2F;&#x2F;xkcd.com&#x2F;505&#x2F;</a> , and abstract it enough times so that you decide that the rocks themselves <i>are not needed anymore</i>.)<p>-&gt; if anyone has any comments about the MUH, would be interested to hear them. My head is kinda-still-aching from the last time I tried to decide how seriously I should take his ideas.
评论 #7696801 未加载
评论 #7696008 未加载
评论 #7696366 未加载
kremlin大约 11 年前
This is a question I&#x27;ve wondered myself. I once read a bit about someone using live crabs to perform computations -- a group of crabs behaves deterministically (enough) to make a series of logic gates that can be used for (very slow) computations. Since reading that, this has been a question on my mind.<p>(source) <a href="http://www.gizmag.com/crab-computer-kobe/22145/" rel="nofollow">http:&#x2F;&#x2F;www.gizmag.com&#x2F;crab-computer-kobe&#x2F;22145&#x2F;</a>
评论 #7695561 未加载
cessor大约 11 年前
In cognitive sciences there is a dispute about the nature of cognition and what scientific tools to apply. Some say cognition is computation, others say its just behavior of a dynamic or connectionist system. A good sumary of aspects can be found in Fresco, Nir, 2012 - The explanatory Role of Computation in Cognitive Science, Minds &amp; Machines, 22, 353-380.<p>The discussion is quite interesting, because computation is very hard to define. Understanding what makes a system compute can yield useful insight for understanding cognition.
评论 #7695633 未加载
csense大约 11 年前
I&#x27;m surprised the authors didn&#x27;t cite the (in)famous &quot;Do Dogs Know Calculus?&quot; article [1] when they ask whether a &quot;dog catching a stick&quot; is an example of computation.<p>[1] PDF link: <a href="http://www.indiana.edu/~jkkteach/Q550" rel="nofollow">http:&#x2F;&#x2F;www.indiana.edu&#x2F;~jkkteach&#x2F;Q550</a>
评论 #7695884 未加载
wyager大约 11 年前
My question:<p>When does a physical system reach a &quot;halting state&quot;?<p>This question has led me to some confusion when considering the possible isomorphism between physical systems and turing machines.
评论 #7695549 未加载
评论 #7695726 未加载
goldenkey大约 11 年前
Furthermore, is time quantized? If it&#x27;s not quantized, are universal changes atomic? Or do they dopple? (At the speed of light?) This would mean that a clock cycle of universal change takes (length of universe) &#x2F; (speed of light)<p>I think that time being quantized, would beg the question, are universal changes atomic locally, or globally?<p>Our perceptions are clearly operating on universal change, so our sense of time is as well. These are quite deep questions.
thisjepisje大约 11 年前
Can&#x27;t we define the universe as the computing system, the state of the universe at t=0 as the encoding of the problem into the computing system, and the state of the universe now as the decoding step? As far as I understand it, the framework presented doesn&#x27;t explicitly permit the computing entity to be inside the computation.
jxjdjr大约 11 年前
I&#x27;ve thought this for a while now. The only reasonable definition of computation is in relation to other systems. It&#x27;s good to see this popping up other places than inside my head :)