4 ms·

Ask HN: What are the best MOOCs for Data Science and Machine Learning?

12 points by bhnmmhmd 9y ago

mycat 9y ago
Yaser Abu-Mostafa's Learning From Data. Theory-heavy tough course but in the long run, really worth it.
nnd 9y ago
I’m currently taking ML course from Andrew Ng, it’s a bit weak in math, but gives you a pretty good understanding how those ML algorithms work.
minimaxir 9y ago
Previous discussion: https://news.ycombinator.com/item?id=14998352 https://news.ycombinator.com/item?id=14998352
iamkeyur 9y ago
Nowadays, there are a couple of really excellent online lectures to get you started. The list is too long to include them all. Every one of the major MOOC sites offers not only one but several good Machine Learning classes, so please check [coursera](https://www.coursera.org/ https://www.coursera.org/), [edX](https://www.edx.org/ https://www.edx.org/), [Udacity](https://www.udacity.com/ https://www.udacity.com/) yourself to see which ones are interesting to you. However, there are a few that stand out, either because they're very popular or are done by people who are famous for their work in ML. Roughly in order from easiest to hardest, those are: * Andrew Ng's [ML-Class at coursera](https://www.coursera.org/course/ml https://www.coursera.org/course/ml): Focused on application of techniques. Easy to understand, but mathematically very shallow. Good for beginners! * Hasti/Tibshirani's [Elements of Statistical Learning](http://statweb.stanford.edu/~tibs/ElemStatLearn/ http://statweb.stanford.edu/~tibs/ElemStatLearn/): Also aimed at beginners and focused more on applications. * Yaser Abu-Mostafa's [Learning From Data](https://www.edx.org/course/caltechx/caltechx-cs1156x-learning-data-1120 https://www.edx.org/course/caltechx/caltechx-cs1156x-learnin...): Focuses a lot more on theory, but also doable for beginners * Geoff Hinton's [Neural Nets for Machine Learning](https://www.coursera.org/course/neuralnets https://www.coursera.org/course/neuralnets): As the title says, this is almost exclusively about Neural Networks. * Hugo Larochelle's [Neural Net lectures](http://www.youtube.com/playlist?list=PL6Xpj9I5qXYEcOhn7TqghAJ6NAPrNmUBH http://www.youtube.com/playlist?list=PL6Xpj9I5qXYEcOhn7TqghA...): Again mostly on Neural Nets, with a focus on Deep Learning * Daphne Koller's [Probabilistic Graphical Models](https://www.coursera.org/course/pgm https://www.coursera.org/course/pgm) Is a very challenging class, but has a lot of good material that few of the other.
socratees 9y ago
I consider Andrew Ng's ML course - https://www.coursera.org/learn/machine-learning https://www.coursera.org/learn/machine-learning the best place to start, in addition to his https://deeplearning.ai https://deeplearning.ai courses.

12 points by bhnmmhmd 9y ago

Yaser Abu-Mostafa's Learning From Data. Theory-heavy tough course but in the long run, really worth it.

nnd 9y ago

I’m currently taking ML course from Andrew Ng, it’s a bit weak in math, but gives you a pretty good understanding how those ML algorithms work.

minimaxir 9y ago

Previous discussion: https://news.ycombinator.com/item?id=14998352 https://news.ycombinator.com/item?id=14998352

iamkeyur 9y ago

Nowadays, there are a couple of really excellent online lectures to get you started. The list is too long to include them all. Every one of the major MOOC sites offers not only one but several good Machine Learning classes, so please check [coursera](https://www.coursera.org/ https://www.coursera.org/), [edX](https://www.edx.org/ https://www.edx.org/), [Udacity](https://www.udacity.com/ https://www.udacity.com/) yourself to see which ones are interesting to you. However, there are a few that stand out, either because they're very popular or are done by people who are famous for their work in ML. Roughly in order from easiest to hardest, those are: * Andrew Ng's [ML-Class at coursera](https://www.coursera.org/course/ml https://www.coursera.org/course/ml): Focused on application of techniques. Easy to understand, but mathematically very shallow. Good for beginners! * Hasti/Tibshirani's [Elements of Statistical Learning](http://statweb.stanford.edu/~tibs/ElemStatLearn/ http://statweb.stanford.edu/~tibs/ElemStatLearn/): Also aimed at beginners and focused more on applications. * Yaser Abu-Mostafa's [Learning From Data](https://www.edx.org/course/caltechx/caltechx-cs1156x-learning-data-1120 https://www.edx.org/course/caltechx/caltechx-cs1156x-learnin...): Focuses a lot more on theory, but also doable for beginners * Geoff Hinton's [Neural Nets for Machine Learning](https://www.coursera.org/course/neuralnets https://www.coursera.org/course/neuralnets): As the title says, this is almost exclusively about Neural Networks. * Hugo Larochelle's [Neural Net lectures](http://www.youtube.com/playlist?list=PL6Xpj9I5qXYEcOhn7TqghAJ6NAPrNmUBH http://www.youtube.com/playlist?list=PL6Xpj9I5qXYEcOhn7TqghA...): Again mostly on Neural Nets, with a focus on Deep Learning * Daphne Koller's [Probabilistic Graphical Models](https://www.coursera.org/course/pgm https://www.coursera.org/course/pgm) Is a very challenging class, but has a lot of good material that few of the other.

socratees 9y ago

I consider Andrew Ng's ML course - https://www.coursera.org/learn/machine-learning https://www.coursera.org/learn/machine-learning the best place to start, in addition to his https://deeplearning.ai https://deeplearning.ai courses.