Re:Seeing Theory: A visual introduction to probability and s...
by sp4ke on Wednesday March 01, 2017 @04:12AM
(#13762395)
As someone learning ML thank you so much ! I hit wall trying to deeply understand the concepts of statistical inference and their relation to Neural Networks and other ML tools.
As I kept reading papers and heavy math material I was hoping an "explorable explanation" existed for this, and I even considered doing it myself. It seems I don't have to anymore. Your series is exactly what was missing.
Not sure if OP is the author but if I may ask, how long did it take you to create all this content ?
Re:Seeing Theory: A visual introduction to probability and s...
by leni536 on Wednesday March 01, 2017 @10:57AM
(#13764707)
The "theoretical" curve on the CLT page at low sample count (like 2) doesn't seem right. I know that CLT is pretty strong, it doesn't converge this fast though. I love the visualizations though, really nice site.
- Ordinary Least Squares Regression
- Principal Component Analysis
- Image Kernels
- Eigenvectors and Eigenvalues
- Markov Chains
- etc
And if you want to learn a bit of D3.js, I collected some materials here: http://p.migdal.pl/2016/02/09/d3js-icm-kfnrd.html
As I kept reading papers and heavy math material I was hoping an "explorable explanation" existed for this, and I even considered doing it myself. It seems I don't have to anymore. Your series is exactly what was missing.
Not sure if OP is the author but if I may ask, how long did it take you to create all this content ?
Thanks again for this beautiful resource.
http://students.brown.edu/seeing-theory/distributions/index....
http://imgur.com/a/MN6zv