Machine Learning Course Materials
cs229.stanford.edu
Machine Learning Course Materials
1–10 of 14 posts
Re: Machine Learning Course Materials
#2Stanford's CS229 was also on Coursera.
Re: Machine Learning Course Materials
#3Stanford's CS229 was also on Coursera. https://class.coursera.org/ml/lecture/preview
Actually, the Courersa class is closer to another course taught by Ng: CS229A - Applied Machine Learning. CS229 is far more theoretical.
Re: Machine Learning Course Materials
#4Stanford's CS229 was also on Coursera. https://class.coursera.org/ml/lecture/preview
I've posted about this many times, but it's worth noting that the Machine Learning course on Coursera is CS229A(plied), which is different from CS229 the one most Stanford students take. The Coursera one is more useful if you want to apply machine learning; the Stanford one linked by the poster is more useful if you want to enter the field. The Coursera one glosses over a lot of the mathematics behind the algorithms; the Stanford one delves into the mathematics supporting the algorithms.
Here are the complete corse materials (including video lectures!) for CS229, courtesy of Stanford Engineering Everywhere: http://see.stanford.edu/see/courseinfo.aspx?coll=348ca38a-3a...
Re: Machine Learning Course Materials
#5Stanford's CS229 was also on Coursera. https://class.coursera.org/ml/lecture/preview
I've posted about this many times, but it's worth noting that the Machine Learning course on Coursera is CS229A(plied), which is different from CS229 the one most Stanford students take. The Coursera one is more useful if you want to apply machine learning; the Stanford one linked by the poster is more useful if you want to enter the field. The Coursera one glosses over a lot of the mathematics behind the algorithms;…
i did take the cs-229A when it was first offered on coursera, and your comment is exactly right. most of the mathematics was glossed over, and the focus was on not "worrying" about it, but on implementation (of various algorithms) using octave, and observe the results. which for neural-nets mostly boiled down to some slightly complicated matrix-multiplication.
having said that, it was an excellent overview of a broad spectrum of ml techniques, and i for one, would heartily recommend it to anyone with high-school background in maths, and a deep interest in the field. supplementing the material with simon-haykin's text or christopher-bishop's (most excellent) book, would make it slightly tougher.
unfortunately, due to some time constraints i could not partake on geoff-hinton's wisdom on deep-learning. would you happen to have material for that stashed somewhere ?
Re: Machine Learning Course Materials
#6Earlier quoted context omitted.
I've posted about this many times, but it's worth noting that the Machine Learning course on Coursera is CS229A(plied), which is different from CS229 the one most Stanford students take. The Coursera one is more useful if you want to apply machine learning; the Stanford one linked by the poster is more useful if you want to enter the field. The Coursera one glosses over a lot of the mathematics behind the algorithms;…
i did take the cs-229A when it was first offered on coursera, and your comment is exactly right. most of the mathematics was glossed over, and the focus was on not "worrying" about it, but on implementation (of various algorithms) using octave, and observe the results. which for neural-nets mostly boiled down to some slightly complicated matrix-multiplication. having said that, it was an excellent overview of a broad…
Ooh, Matrix multiplication? Would I get to use Strassen's sub-cubic algorithm? I've been looking for an excuse to use it.
Re: Machine Learning Course Materials
#7Earlier quoted context omitted.
I've posted about this many times, but it's worth noting that the Machine Learning course on Coursera is CS229A(plied), which is different from CS229 the one most Stanford students take. The Coursera one is more useful if you want to apply machine learning; the Stanford one linked by the poster is more useful if you want to enter the field. The Coursera one glosses over a lot of the mathematics behind the algorithms;…
i did take the cs-229A when it was first offered on coursera, and your comment is exactly right. most of the mathematics was glossed over, and the focus was on not "worrying" about it, but on implementation (of various algorithms) using octave, and observe the results. which for neural-nets mostly boiled down to some slightly complicated matrix-multiplication. having said that, it was an excellent overview of a broad…
I've not had the time to go through the Hinton course fully myself, but the Coursera link is still up - https://www.coursera.org/course/neuralnets
You should still be able to register/download the materials. That's what I did.
Re: Machine Learning Course Materials
#8I wrote a blog post with some material suggestions to learn Machine Learning. If you want, visit my blog post:
http://pauloortins.com/resources-to-become-a-ninja-machine-l...
Re: Machine Learning Course Materials
#9Earlier quoted context omitted.
i did take the cs-229A when it was first offered on coursera, and your comment is exactly right. most of the mathematics was glossed over, and the focus was on not "worrying" about it, but on implementation (of various algorithms) using octave, and observe the results. which for neural-nets mostly boiled down to some slightly complicated matrix-multiplication. having said that, it was an excellent overview of a broad…
Ooh, Matrix multiplication? Would I get to use Strassen's sub-cubic algorithm? I've been looking for an excuse to use it.
afaik, octave uses blas, which in turn should be using strassen's for it's sgemm, dgemm etc. computations. at least it would be very surprising if it didn't...
Re: Machine Learning Course Materials
#10Stanford's CS229 was also on Coursera. https://class.coursera.org/ml/lecture/preview
I did that course and it's very different. They are complementary, the one in Coursera being more introductory.
I'd recommend the Caltech course too, which is also tougher.