colah / colah.github.io Goto Github PK
View Code? Open in Web Editor NEWHome Page: colah.github.io
Home Page: colah.github.io
Hi! I noticed you have an RSS feed at https://colah.github.io/rss.xml, but it's not linked from anywhere. Please add the following to <head>
of each page so that newsreaders could find the feed automatically:
<link rel="alternate" type="application/rss+xml" title="colah's blog" href="rss.xml">
Thank you.
First off, I would just like to say that your blog posts are awesome. The math you are blogging about is extremely complicated (if you are trying to read it via papers) but your posts definitely explain it in Layman's terms better than anything I have come across.
I would change the wording this paragraph from:
"At its most basic, convolutional neural networks can be thought of as a kind of neural network that uses many identical copies of the same neuron.1 This allows the network to have lots of neurons and express computationally large models while keeping the number of actual parameters – the values describing how neurons behave – that need to be learned fairly small."
to something like this
"At its most basic, convolutional neural networks can be thought of as a kind of neural network that uses many identical copies of the same neuron.1 This allows the network to have lots of neurons and express computationally large models while keeping the number of actual parameters (the values describing how neurons behave) that have to be learned fairly small."
I had to read that last sentence a few times to figure out what was going on. This makes more sense to me. Keep up the good work!
Hello Mr. Olah,
I was going through your blog and really liked the design of the same. I was wondering if you are able to point me to how you set up the same - perhaps a website with instructions or if I can download your repo and try to recreate the same.
Thanks,
RJaikanth
https://camo.githubusercontent.com/90667bfb975db1e7ab28e6a176e722a9820205b2c3d0a912fae312b15477b46f/68747470733a2f2f636f6c61682e6769746875622e696f2f706f7374732f323031352d30382d556e6465727374616e64696e672d4c53544d732f696d672f4c53544d332d7661722d4752552e706e67
The picture you drawed is very fancy. can I use it to introduce RNN in my paper ? Thank you very much.
colah.github.io has no automatic redirection from HTTP to HTTPS. It should, but until then, please make explicit calls to HTTPS.
Hi, I think the lstm diagrams in your blog are fancy, may I ask what tool did you use to draw them?
The hidden state is never described as part of LSTM, and is only mentioned in the section about GRU.
There is a paragraph that refers to both the cell state (C_t) and the hidden state (h_t) as "cell state". This is quite confusing, although the nearby graph provides some insight to what is really meant.
In Understanding LSTM Networks blog the LSTM diagram has wrong labels this & this image. These are the correct labels:
Please can you tell me how you made your figures?! Particularly in this post and very specifically this image. Are you using Tikz or something manual?
Thank you for the great blogs.
Hello, Mr. Olah.
First thing first, thank you for your great work. I am learning with pleasure via your articles.
Here I would like to address what is difficult for me to understand.
The below two sentences are from the source of the section "Shared Representations":
This general tactic ... One of the great strengths of this approach is that it allows the representation to learn from more than one kind of data.
There’s a counterpart to this trick. Instead of learning a way to represent one kind of data and using it to perform multiple kinds of tasks, we can learn a way to map multiple kinds of data into a single representation!
It was kind of confusing for me, because "the trick" and its "counterpart" seem to have the similar descriptions. And below is my current understanding.
For the first one, the representation can learn other kind of data, incrementally, whenever other tasks come in and the word embedding tries to solve the tasks. And for the second one (its counterpart), assuming we already have multiple tasks and its corresponding multiple kinds of data, then the single representation can learn these all data.
Since I have worked this article for the translation in Korean, it would be great if my understanding gets confirmed and corrected.
Best regards,
Don
A declarative, efficient, and flexible JavaScript library for building user interfaces.
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google ❤️ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.