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。

Failures of Deep Learning

210 点作者 stochastician大约 8 年前

5 条评论

Houshalter大约 8 年前
Just yesterday there was a big discussion on here about how academic papers needlessly complicate simple ideas. Mainly by replacing nice explanations with impenetrable math notation in an attempt to seem more formal. This paper is very guilty of this.<p>E.g. page 5. They attempt to explain a really simple idea, that they generated images of random lines at a random angle. Then labelled the lines positive or negative examples, based on whether the angle was greater than 90 degrees or not. Then they take sets of these examples. And label them based on whether they contain an even or odd number of positive examples.<p>They take several paragraphs over half a page to explain this. Filled with dense mathematical notation. If you don&#x27;t know what symbols like U, :, -&gt;, or ~ mean, you are screwed because that&#x27;s not googleable. It takes way longer to parse than it should. Especially since I just wanted to quickly skim the ideas, not painfully reverse engineer them.<p>Hell, even <i>the concept of even or odd</i>, is pointlessly redefined and complicated as multiplying + or - 1&#x27;s together. I was scratching my head for a few minutes just trying to figure out what the purpose of that was. It&#x27;s like reading bad code without any comments. Even if you are very familiar with the language and know what the code does, it takes a lot of effort to figure out <i>why</i> it&#x27;s that way. If it&#x27;s not explained properly.<p>The worst part is, no one ever complains about this stuff because they are afraid of looking stupid. I sure fear that by posting this very comment. I actually am familiar with the notation used in this example. I still find it unnecessary and exhausting to decode.
评论 #13954614 未加载
评论 #13953577 未加载
评论 #13957116 未加载
评论 #13955334 未加载
评论 #13954667 未加载
评论 #13958017 未加载
评论 #13956778 未加载
评论 #13954741 未加载
评论 #13954425 未加载
评论 #13953822 未加载
dmreedy大约 8 年前
I think some of the most exciting and interesting work comes out of proving, not just capabilities, but constraints for systems, be it Gödel, Shannon, Aaronson, or any of the others in the smaller-than-desirable tradition of those who say, &quot;No&quot;. I think a better understanding what Deep Learning <i>can&#x27;t</i> do (well) is fertile material for better understanding the kinds of problems it <i>can</i> do, and am very excited to see more work in this space, and movement towards an underlying structural theory.
评论 #13952589 未加载
评论 #13952256 未加载
评论 #13952953 未加载
csfoo大约 8 年前
The lead author will be giving a talk on this work next week (which will be live streamed and recorded) as part of a workshop on Representation Learning:<p><a href="https:&#x2F;&#x2F;simons.berkeley.edu&#x2F;talks&#x2F;shai-shalev-shwartz-2017-3-28" rel="nofollow">https:&#x2F;&#x2F;simons.berkeley.edu&#x2F;talks&#x2F;shai-shalev-shwartz-2017-3...</a>
smdz大约 8 年前
Link to the PDF: <a href="https:&#x2F;&#x2F;arxiv.org&#x2F;pdf&#x2F;1703.07950.pdf" rel="nofollow">https:&#x2F;&#x2F;arxiv.org&#x2F;pdf&#x2F;1703.07950.pdf</a>
bra-ket大约 8 年前
the biggest failure of deep learning is the lack of common sense
评论 #13956233 未加载