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。

Classical sorting algorithms as a model of morphogenesis (2023)

166 点作者 filoeleven5 个月前

9 条评论

mithametacs5 个月前
Building something that <i>isn&#x27;t</i> Turing-complete is surprisingly-hard once it&#x27;s complex enough.<p>If basal intelligence is present in diverse computational structures, then weak intelligence is everywhere.<p>If weak intelligence is everywhere, Earth-like planets are everywhere, ... where are the aliens?<p>Personally, I blame game theory. Too many agents too smart in one place, you get conflicts, and eventually someone breaks an atom apart in your direction.<p>Or do you need emotions to have conflict? Are there basal emotions?<p>I&#x27;m usually not worried about AI uprisings, but I do believe in the possibility of conflict.
评论 #42459772 未加载
评论 #42461221 未加载
评论 #42460275 未加载
评论 #42458576 未加载
评论 #42458361 未加载
评论 #42458077 未加载
评论 #42458882 未加载
评论 #42461140 未加载
评论 #42459266 未加载
评论 #42458969 未加载
jstrieb5 个月前
The last coauthor listed on this preprint is Michael Levin, who has a lot of other cool work.<p>In particular, this talk of his from NeurIPS 2018 includes fascinating biology research results, as well as musings on the future of biologically-inspired artificial intelligence.<p><a href="https:&#x2F;&#x2F;youtu.be&#x2F;RjD1aLm4Thg" rel="nofollow">https:&#x2F;&#x2F;youtu.be&#x2F;RjD1aLm4Thg</a><p>HN discussion about the talk: <a href="https:&#x2F;&#x2F;news.ycombinator.com&#x2F;item?id=18736698">https:&#x2F;&#x2F;news.ycombinator.com&#x2F;item?id=18736698</a>
评论 #42463042 未加载
评论 #42459114 未加载
评论 #42458603 未加载
sitkack5 个月前
&quot;The Collective Intelligence of Morphogenesis: a model system for basal cognition&quot; by Michael Levin<p><a href="https:&#x2F;&#x2F;www.youtube.com&#x2F;watch?v=JAQFO4g7UY8" rel="nofollow">https:&#x2F;&#x2F;www.youtube.com&#x2F;watch?v=JAQFO4g7UY8</a><p>And from Machine Learning Street Talk<p>Michael Levin - Why Intelligence Isn&#x27;t Limited To Brains.<p><a href="https:&#x2F;&#x2F;www.youtube.com&#x2F;watch?v=6w5xr8BYV8M" rel="nofollow">https:&#x2F;&#x2F;www.youtube.com&#x2F;watch?v=6w5xr8BYV8M</a>
评论 #42466578 未加载
danwills5 个月前
I loved this paper, below is the punchline for me I reckon (and summarizes where-they-are-at nicely too I think - finding out a lot of possibly very important things!):<p>&quot;&quot;The discovery of unexpected problem-solving competencies (such as Delayed Gratification and segregation) that are not apparent from the component policies and algorithms themselves is a critical research program.&quot;&quot;
niemandhier5 个月前
This is fun, but I would call it an example of self organization &#x2F; self organized complexity not intelligence.<p>Cell membranes assemble themselves, so do micella ( little spherical protein baubles ), or to take a non living example lipid bilayers.<p>We would not call such a system intelligent.
评论 #42460294 未加载
评论 #42479160 未加载
danwills5 个月前
I just loved this bit in the paper, could be so easily taken on so many tangents!:<p>&quot;&quot;Delayed Gratification is used to evaluate the ability of each algorithm undertake actions that temporarily increase Monotonicity Error in order to achieve gains later on. Delayed Gratification is defined as the improvement in Sortedness made by a temporarily error-increasing action.&quot;&quot;<p>Is it slightly analogous in some ways to the avoidance of getting stuck in local maxima perhaps?
评论 #42460466 未加载
ziofill5 个月前
I recommend Michael Levin’s YouTube channel. Lots and lots of fascinating discussions.
InDubioProRubio5 个月前
A beach sorts itself by size of sand, just by applying physics, so any array copied by parallel processes will sort itself given enough time, by computation time spend on the element aka the size of the rocks.
评论 #42460396 未加载
trenchgun5 个月前
Are these just traditional distributed algorithms?