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.

Pen and Paper Exercises in Machine Learning (2022)

413 pointsby ibobev2 months ago

11 comments

lucasoshiro2 months ago
Seems to be cool, but, one of thing that most annoys me on studying machine learning is that I may dive as deep as it is possible in theory, but I can&#x27;t see how it connects to the practice, i. e. how it makes me choose the correct number of neurons in a layer, how many layers, the activation functions, if I should use a neural network or other techniques, and so on...<p>If someone have something explaining that I&#x27;ll be grateful
评论 #43441567 未加载
评论 #43441582 未加载
评论 #43441245 未加载
评论 #43444883 未加载
评论 #43441858 未加载
评论 #43441197 未加载
评论 #43454473 未加载
评论 #43447621 未加载
评论 #43442631 未加载
评论 #43445830 未加载
评论 #43451360 未加载
评论 #43444714 未加载
评论 #43444054 未加载
评论 #43441354 未加载
simojo2 months ago
Very neat! Reminds me of Tom Yeh&#x27;s &quot;AI By Hand&quot; exercises [0].<p>[0] <a href="https:&#x2F;&#x2F;www.byhand.ai&#x2F;" rel="nofollow">https:&#x2F;&#x2F;www.byhand.ai&#x2F;</a>
评论 #43441023 未加载
S4M2 months ago
Looks neat! My only criticism would be that the solutions are given right after the questions so I couldn&#x27;t help to read the answer of one question before thinking it through by myself.
plants2 months ago
This is really neat! I work in machine learning but still feel imposter syndrome with my foundations with math (specifically linear algebra and matrix&#x2F;tensor operations). Does anyone have any more good resources for problem sets with an emphasis on deep learning foundational skills? I find I learn best if I do a bit of hands-on work every day (and if I can learn things from multiple teachers’ perspectives)
antipaul2 months ago
So who among current ML practitioners building “useful” ML could solve some of these?<p>_Should they_ be able to?
评论 #43440657 未加载
评论 #43441027 未加载
评论 #43440698 未加载
评论 #43443818 未加载
评论 #43441049 未加载
kingkongjaffa2 months ago
complete with solutions, beautiful, thank you for sharing!<p>I&#x27;d be interested in more of these pen and paper exercises, if there is such a term, for other topics.
评论 #43440742 未加载
dang2 months ago
Discussed at the time:<p><i>Pen and paper exercises in machine learning (2021)</i> - <a href="https:&#x2F;&#x2F;news.ycombinator.com&#x2F;item?id=31913057">https:&#x2F;&#x2F;news.ycombinator.com&#x2F;item?id=31913057</a> - June 2022 (55 comments)
FilosofumRex2 months ago
Funny how mathematicians always try to sneak their linear algebra and matrix theory into ML. If you didn&#x27;t know any better, you&#x27;d think academicians had invented LLMs and are the experts to be consulted with.<p>If anything academicians and theoreticians held ML back and forced generations of grad students doing symbolic proofs, like in this example, just because computational techniques were too lowbrow for them.
评论 #43443808 未加载
评论 #43443407 未加载
评论 #43443296 未加载
BeetleB2 months ago
Isn&#x27;t arxiv meant for research level papers? Surprised to see this hosted there.
axpy9062 months ago
Love it.
imranq2 months ago
If someone could turn these into an adaptive Khan Academy style app, that would be incredible
评论 #43441951 未加载