Norm Matloff
matloff.wordpress.com · Personal blog · on HN since 2014
Norm Matloff, a statistics and computer science professor, writes about R, statistics, data science, and why neural networks resemble polynomial regression.
What HN saysHN links it mainly for his polynomial regression argument about neural networks; one commenter has a low opinion of him, and the paper was revised after criticism.
19HN people shared it
7linked it in comments
14stories submitted
464story points
4 years agolast shared
2014people per year2026
Most-shared posts
- 1Neural Networks Are Essentially Polynomial Regression (2018)5 people · 295 points · first shared 2019 · 75 comments on HN
- 2R beats Python, R beats Julia, Anyone else wanna challenge R?3 people · 81 points · first shared 2014 · 49 comments on HN
- 3Statistics: Losing Ground to CS, Losing Image Among Students3 people · 18 points · first shared 2014 · 3 comments on HN
- 4Simpson's Paradox is Back1 people · 56 points · first shared 2014 · 17 comments on HN
- 5Some Comments on Donaho’s “50 Years of Data Science”1 people · 4 points · first shared 2016 · discussion on HN
- 6Neural Networks Are Essentially Polynomial Regression1 people · 3 points · first shared 2018 · discussion on HN
- 7Partools: A sensible R package for large data sets1 people · 2 points · first shared 2015 · discussion on HN
- 8Latest on the Julia Language (vs. R)1 people · 2 points · first shared 2016 · discussion on HN
- 9Why do DL networks work well in spite of being drastically overparameterized?1 people · 2 points · first shared 2021 · discussion on HN
- 10Comments on the New R OOP System, R71 people · 1 points · first shared 2022 · discussion on HN
- 11Update on polynomial regression in lieu of neural nets1 people · first shared 2019
- 12What can go wrong my favorite example1 people · first shared 2014
What HN people wrote when they linked it
tokenadult · 2014-05-01
This blog post by the same author looks pretty good too. "What Can Go Wrong: My Favorite Example" (27 April 2014)
http://matloff.wordpress.com/2014/04/27/what-can-go-wrong-my-favorite-…
jonnycomputer · 2021-10-12
Incidentally, yesterday I submitted a wordpress blog post about some of these issues.
https://news.ycombinator.com/item?id=28817666
https://matloff.wordpress.com/2020/11/11/the-notion-of-double-descent/
bigbillheck · 2021-05-24
I've had a low opinion of Matloff since at least https://matloff.wordpress.com/2018/06/20/neural-networks-are-essential…
thesz · 2020-11-24
> Also one could argue that the polynomial model is just a perceptron
One also can argue otherwise [1].
[1] https://matloff.wordpress.com/2018/06/20/neural-networks-are-essential…
carapace · 2019-08-30
They revised the paper in response to earlier, uh, response.
https://matloff.wordpress.com/2018/07/01/update-on-polynomial-regressi…
https://arxiv.org/abs/1806.06850
kgwgk · 2019-08-30
The actual blog post: https://matloff.wordpress.com/2018/06/20/neural-networks-are-essential…
baldfat · 2015-03-18
in R loops are generally frown upon due to this issue.
BUT R is not slow when it comes to parallel processing or using Revolution Anayltics also has speed ups if you need it. R also has dplyr is speedy and the data.tables is even faster. I think the original Julia speed claims were a little biased to Julia and well there is plenty of awesome things about R, but "slow" isn't a far statement. There is a reason why R has grown so much.
Interesting argument for R usage:
https://matloff.wordpress.com/2014/05/21/r-beats-python-r-beats-julia-…