The Best Advice From Quora on ‘How to Learn Machine Learning’
Top machine learning writers on Quora give their advice on learning machine learning, including specific resources, quotes, and personal insights, along with some extra nuggets of information.
By Matthew Mayo.
How do I learn machine learning?
Straightforward question. Not-so-straightforward answer.
There are obviously a number of ways to go about learning machine learning, with books, courses, and degree programs all being great places to start. Like any number of topics a newcomer may delve into, however, there are a vast number of options in each of these categories, and attempting to narrow one's focus alone can often prove futile.
A Quora post, aptly titled 'How Do I Learn Machine Learning?,' ends up being a robust resource. The FAQ has generated a lot of attention during the course of its life, with 93 answers and more than 468,000 views, and has contributions from a number of well-known personalities in the machine learning world. The idea of learning from others whom have previously undertaken the same task has special significance for the learning of machine learning.
In this post we will take a look at advice from the top answers of the Quora post. We will find recommended courses and books relevant to learning machine learning, garner specific advice from experts, and see what other nuggets we can pick up along the way.
As you look over this advice, don't forget to check our other fine learning material and resources, including a curated list of free data science books, collections of top data science and mathematics MOOCs, and a large number of advice posts and tutorials. Also keep in mind that, in contrast with many of the other educational resources on this site, this post deals solely with machine learning, and while a prominent component of data science, this advice is more narrowly-focused than some of the other data science learning materials.
Our Advisors
Our advisors today are the authors of the 3 most-upvoted FAQ answers, and come in the form of 3 well-known machine learning personalities:
Taken together, our advisors' recommendations compose a strong collection of introductory texts, covering statistical learning, the theoretical underpinnings of machine learning, and the practical implementation of algorithms and model-building in the most popular programming languages (Python & R) and framework (Spark).
Xavier recommends:
Sean recommends:
It's nearly unanimous in most circles which machine learning MOOC is best for newcomers: Andrew Ng's Coursera offering. Beyond that, 2 other Coursera courses are also given specific mention. Incidentally, all 3 MOOC recommendations come from Xavier, with Sean co-signing the Ng selection.
How do I learn machine learning?
Straightforward question. Not-so-straightforward answer.
There are obviously a number of ways to go about learning machine learning, with books, courses, and degree programs all being great places to start. Like any number of topics a newcomer may delve into, however, there are a vast number of options in each of these categories, and attempting to narrow one's focus alone can often prove futile.
A Quora post, aptly titled 'How Do I Learn Machine Learning?,' ends up being a robust resource. The FAQ has generated a lot of attention during the course of its life, with 93 answers and more than 468,000 views, and has contributions from a number of well-known personalities in the machine learning world. The idea of learning from others whom have previously undertaken the same task has special significance for the learning of machine learning.
In this post we will take a look at advice from the top answers of the Quora post. We will find recommended courses and books relevant to learning machine learning, garner specific advice from experts, and see what other nuggets we can pick up along the way.
As you look over this advice, don't forget to check our other fine learning material and resources, including a curated list of free data science books, collections of top data science and mathematics MOOCs, and a large number of advice posts and tutorials. Also keep in mind that, in contrast with many of the other educational resources on this site, this post deals solely with machine learning, and while a prominent component of data science, this advice is more narrowly-focused than some of the other data science learning materials.
Our Advisors
Our advisors today are the authors of the 3 most-upvoted FAQ answers, and come in the form of 3 well-known machine learning personalities:
- Xavier Amatriain, VP of Engineering at Quora
- Raviteja Chirala, Data Scientist at Ayasdi
- Sean McClure, Senior Data Scientist at ThoughtWorks
Taken together, our advisors' recommendations compose a strong collection of introductory texts, covering statistical learning, the theoretical underpinnings of machine learning, and the practical implementation of algorithms and model-building in the most popular programming languages (Python & R) and framework (Spark).
Xavier recommends:
- Hastie, Tibshirani, and Friedman's The Elements of Statistical Learning
- Bishop's Pattern Recognition and Machine Learning
- David Barber's Bayesian Reasoning and Machine Learning
- Kevin Murphy's Machine learning: a Probabilistic Perspective
- Larry Wasserman's All of Statistics: A Concise Course in Statistical Inference
Sean recommends:
- Brett Lantz's Machine Learning with R
- Willi Richert & Luis Pedro Coelho's Building Machine Learning Systems with Python
- Nick Pentreath's Machine Learning with Spark
- Yaser Abu-Mostafa, Malik Magon-Ismail, Hsuan-Tien Lin's Learning from Data
It's nearly unanimous in most circles which machine learning MOOC is best for newcomers: Andrew Ng's Coursera offering. Beyond that, 2 other Coursera courses are also given specific mention. Incidentally, all 3 MOOC recommendations come from Xavier, with Sean co-signing the Ng selection.
- Andrew Ng's Machine Learning [Stanford] via Coursera
- Joseph Konstan & Michael Ekstrand's Introduction to Recommender Systems [U. Minnesota] via Coursera
- Jure Leskovec, Anand Rajaraman & Jeff Ullman's Mining Massive Data Sets [Stanford] via Coursera