Learning Machine Learning: A beginner's journey • Hacker News Board

Re: Learning Machine Learning: A beginner's journey

#41

Post by webmaven » Mon, Dec 26, 2016, 8:43 PM UTC

solipsism wrote: post #38

saurabhjha wrote: post #10

I think this "machine learning for hackers" approach is just not enough. Oftentimes, you do need a solid theoretical/mathematical background. Most people seems to approach ML like they approach programming tools or libraries - learn just enough to get job done and move on. I was studying machine learning from Andrew Ng's CS229 (the class videos are online. I think they date from 2008 or hereabout). There is no way yo…

This is backward thinking. It borders on elitist, although I know it's not meant that way. Developers everywhere use Paxos without even knowing it, much less having read Lamport's papers, because they're building on top of solid tools that use Paxos (or Raft or what have you). This is more true at Google and Facebook than anywhere. Same goes for ML. You can study the theory, and you can learn to apply it. In the fiel…

> eventually robust tools are built upon which developers can build systems without having "studied these things for years (by PhD)".

For ML, I don't think we are at the eventually point just yet.

Re: Learning Machine Learning: A beginner's journey

#42

Post by solipsism » Mon, Dec 26, 2016, 9:29 PM UTC

webmaven wrote: post #41

solipsism wrote: post #38

Earlier quoted context omitted.

This is backward thinking. It borders on elitist, although I know it's not meant that way. Developers everywhere use Paxos without even knowing it, much less having read Lamport's papers, because they're building on top of solid tools that use Paxos (or Raft or what have you). This is more true at Google and Facebook than anywhere. Same goes for ML. You can study the theory, and you can learn to apply it. In the fiel…

> eventually robust tools are built upon which developers can build systems without having "studied these things for years (by PhD)". For ML, I don't think we are at the eventually point just yet.

Let's put it this way. A startup that insists every one of its developers touching ML has a sound basis in fundamental theory is going to get left in the dust.

Anyone who wants to be the guy/gal who understands the fundamentals will be valuable. But we don't need everyone trying to be that person. And most wouldn't be successful, though they'd be successful as the guy/gal who does other stuff.

Re: Learning Machine Learning: A beginner's journey

#43

Post by bronxbomber92 » Mon, Dec 26, 2016, 10:12 PM UTC

chestervonwinch wrote: post #24

Philipp__ wrote: post #22

Distributed Systems and ML are probably two most interesting things that I have on the radar, that got me really scared to the point where I do not know from where to start, and most importantly for what?! Most of my free time (time I spent on personal projects) was writing physics simulation in Java, playing with Lisp and doing some backend development. Nothing amazing. Year and a half ago I got really interested in…

> If I choose to go Ai route, I do not know from where to start ... This type of comment is often made in machine learning (ML) related submissions. The pre-req list is long: calculus, linear algebra, stats, probability, numerical methods (for optimization, linear algebra, maybe interpolation), etc. BUT, you don't really need to go through the entirety of each subject for ML. For example, in calculus, you probably on…

Here's what I've found along the lines of "mathematics for machine learning":

* DS-GA 1002: Statistical and Mathematical Methods (http://www.cims.nyu.edu/~cfgranda/pages/DSGA1002_fall15/inde...) by Carlos Fernandez-Granda of NYU

(There is a 2016 version of the course with different lectures notes as well.)

* Numerical Algorithms (http://people.csail.mit.edu/jsolomon/share/book/numerical_bo...) by Justin Solomon

* Math for Intelligent Systems, 2016 (https://ipvs.informatik.uni-stuttgart.de/mlr/teaching/maths-...) by Marc Toussaint & Hung Ngo of University Stuttgart

* Math for Intelligent Systems, 2015 (https://ipvs.informatik.uni-stuttgart.de/mlr/teaching/mathem...) by Nathan Ratliff of University Stuttgart

* Mathematics for Inference and Machine Learning (http://wp.doc.ic.ac.uk/sml/teaching/mathematics-for-machine-...) by Stefanos Zafeiriou and Marc Deisenroth of Imperial College London

Most of these are lectures notes. Some are very detailed, but I think there is still space for a completely fleshed out book on the subject.

Re: Learning Machine Learning: A beginner's journey

#44

Post by webmaven » Tue, Dec 27, 2016, 4:36 AM UTC

solipsism wrote: post #42

webmaven wrote: post #41

Earlier quoted context omitted.

> eventually robust tools are built upon which developers can build systems without having "studied these things for years (by PhD)". For ML, I don't think we are at the eventually point just yet.

Let's put it this way. A startup that insists every one of its developers touching ML has a sound basis in fundamental theory is going to get left in the dust. Anyone who wants to be the guy/gal who understands the fundamentals will be valuable. But we don't need everyone trying to be that person. And most wouldn't be successful, though they'd be successful as the guy/gal who does other stuff .

Well, sure. But that isn't "eventually".

Re: Learning Machine Learning: A beginner's journey

#45

Post by nercht12 » Tue, Dec 27, 2016, 5:10 AM UTC

stevehiehn wrote: post #34

saurabhjha wrote: post #10

I think this "machine learning for hackers" approach is just not enough. Oftentimes, you do need a solid theoretical/mathematical background. Most people seems to approach ML like they approach programming tools or libraries - learn just enough to get job done and move on. I was studying machine learning from Andrew Ng's CS229 (the class videos are online. I think they date from 2008 or hereabout). There is no way yo…

Everytime there is a paradigm shift there is always that voice: If you don't understand the paint at a chemical compound level you can't make a beautiful painting. Wait what?

Eh, let's revise that analogy. More like not understanding the bricks means you can't make a good building. You can by good intuition, but it won't be spot-on perfect (as by calculating all the physics) and you'll need more luck the higher you get.

Re: Learning Machine Learning: A beginner's journey

#46

Post by jczhang » Tue, Dec 27, 2016, 9:08 AM UTC

pramodliv1 wrote: post #28

ken47 wrote: post #12

Earlier quoted context omitted.

I don't think Andrew Ng would agree with your assertion. His Coursera ML class assumes little more than a basic high school math education, and at the start of the course, he teaches the very small subset of linear algebra required to understand his course materials. I think what Andrew Ng would say is that without a rigorous statistical background, you will be limited in your ability to use ML, and you will certainl…

> And at the start of the course, he teaches the very small subset of linear algebra required to understand his course materials. I tried doing the ML course without any prior knowledge of Linear Algebra and dropped out after the first three weeks. In hindsight, I realized it wouldn't have been possible to appreciate how PCA works without understanding eigenvectors, how collaborative filtering is an elegant applicati…

Would you recommend that Linear Algebra course?

Re: Learning Machine Learning: A beginner's journey

#47

Post by aspiringme » Tue, Dec 27, 2016, 9:16 AM UTC

Machine learning.. is the new avenue mankind can boast of.

Re: Learning Machine Learning: A beginner's journey

#48

Post by tuxguy » Tue, Dec 27, 2016, 9:49 AM UTC

saurabhjha wrote: post #10

I think this "machine learning for hackers" approach is just not enough. Oftentimes, you do need a solid theoretical/mathematical background. Most people seems to approach ML like they approach programming tools or libraries - learn just enough to get job done and move on. I was studying machine learning from Andrew Ng's CS229 (the class videos are online. I think they date from 2008 or hereabout). There is no way yo…

> Well, I think a lot of competitive advantage of Google/Facebook in ML is because they have staffed their engineering with people who have studied these things for years (by PhD). Compare that to flipkart's recommendations.

Not entirely true. Google & FB have orders of magnitude more data than flipkart. You can have the smartest ML people on the planet churning out the most clever, advanced ML algos & models, but without enough data, its not going to be useful & effective.

I recently attended slashn[0], flipkart's annual technical conference, and spoke to a bunch of their ML folks. They have masters & phd degrees in ML from IITs, IISc and are as smart as they come.

Sure, flipkart doesnt have marquee names of the likes of yann lecun, andrew ng, but i wouldnt doubt the ML talent Flipkart has

0. https://slashn.flipkart.net/ ps : I dont work for flipkart, but have friends who work in ML teams

Re: Learning Machine Learning: A beginner's journey

#49

Post by tuxguy » Tue, Dec 27, 2016, 9:53 AM UTC

saurabhjha wrote: post #10

I think this "machine learning for hackers" approach is just not enough. Oftentimes, you do need a solid theoretical/mathematical background. Most people seems to approach ML like they approach programming tools or libraries - learn just enough to get job done and move on. I was studying machine learning from Andrew Ng's CS229 (the class videos are online. I think they date from 2008 or hereabout). There is no way yo…

Love the statistics recommendations by Prof. Jordan ! Thanks a bunch !

Re: Learning Machine Learning: A beginner's journey

#50

Post by pramodliv1 » Tue, Dec 27, 2016, 11:14 AM UTC

jczhang wrote: post #46

pramodliv1 wrote: post #28

Earlier quoted context omitted.

> And at the start of the course, he teaches the very small subset of linear algebra required to understand his course materials. I tried doing the ML course without any prior knowledge of Linear Algebra and dropped out after the first three weeks. In hindsight, I realized it wouldn't have been possible to appreciate how PCA works without understanding eigenvectors, how collaborative filtering is an elegant applicati…

Would you recommend that Linear Algebra course?

Sure! The course has a slow start. The first few lectures focus on the mechanics of matrix operations instead of starting with linear transformations, and compositions of linear transformation - the core ideas behind Linear Algebra. But after lecture 10, the course picks up pace.