Lecture 1 | Machine Learning (Stanford) 

thank u sooo much!!!!

Thank you

If this was Reddit, I'd give you gold. You, sir, are a hero.

Thank you so much for your efforts.....

Thank you!

u rite

You're grateful and I on the other hand have no idea what he's saying.

Also, thought this was an Andrew YANG video lol

@@samsonsu1541me neither .. but still we are grateful :)

@@Kakerate2 km nnbnm mmmmmxih

Thanks

Thanks, I felt to comment the same and found your comment prior to that. :)

TheShreester thanks

thanks here is the link to the start ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-UzxYlbK2c7E.htmlm34s

thanks

What will you do with AI at this age?

I’m 19 all older guys hang on, I’m coming

Ssss#sss draw t8euufsysy9ày88y8s8y8sy8stss88888f8fsf8fss88ts88rssrsssssssyy8y88tsssdssttttttttt888888888888 try tttt[[[=tye dye=e|you 6seseeseeetddrrt8[do r9[you know [[tragus edr=ee6rrrrrrr[ztdtsst[you see everyone[eeeeeeeeeeeeeeeeeetett er r r eeeet e trade so far far darts stays ẞtd7a8da y777aay7aa77afffaa to fy] SA

Keep learning, everyday is a chance to gain knowledge

Way to go, Parag!

Ikr ???

saitama is goku stronger than you?

It's ok, i'll only give you half a fault. Sit where you want :).

You are better off....I am six months behind you..Late to the bus yet again..

i am just here to read trolls

Christian Gentry better skip to 33 minutes and save 3 more minutes of your life :)

Охтеров Егор Thx guys ;-)

Christian Gentry thank you man

Have you guys checked out the latest nvidia videos? Pretty amazing how fast this is moving.

Christian Gentry thaaaaaaank you, but i wacthed 29 minutes haha

vigilhammer really? I didn't know

He also was a Co founder in Baidu, China's Google I guess...

He´s not human!;)

@@sandipandas8272 ط(£٠£^€

Thank you!

Yep its pretty heavy stuff unless you are really up to speed with linear algebra and probabilistic theory. So useful to mix it up alongside simple introductory ML books, practice with running algorithms against sample sets in R, Python or Weka. Slowly it starts to sink in?

Depends on if you want to learn ML at a practical level or at an academic level. If you plan on getting good enough at ML to get a job in the field then you'll want to take the practical route, but if you want to learn ML just for the sake of understanding it or doing research then you'll want to go the academic route. Andrew's lectures are highly academic and focus a lot on the mathematical and statistical aspects of ML so unless you have a solid foundation of statistics and linear algebra basics you'll struggle to keep up. Additionally, if you follow his lectures on Coursera then you'll want to learn MATLAB as well as that's the programming language used there. MATLAB is great for prototyping when coming up with your own ML scenarios or algorithms but it's not all that practical for field work (i.e developing an ML application, gathering and managing massive amounts of data, etc.). But the good thing about Andrew's lectures is that he's incredibly well versed in the subject and could explain its concepts much better than almost anyone else I've found online. If you get through one of his courses you'll have a thorough understanding of how most ML algorithms work and that'll form a solid foundation for practical application. However, if you want to study ML for the sake of deploying applications or getting a job then there are other available courses online which focus heavily on getting you to apply ML to real world scenarios as soon as possible. One such series is Google's own newly started Intro to Machine Learning (goo.gl/lIuJb2). They use Python to get you up and running coding your first ML program asap. However, the series is not complete and they are still adding to it at the time of writing. Another great source, in my opinion, is Sentdex's tutorials (goo.gl/RteOHz). He has a large number of tutorial series focused on implementing ML using Python which are aimed at solving practical, everyday problems like stock market predictions. These tutorials mainly focus on using already-available Python libraries (such as Sklearn and MatPlotLib) to implement ML algorithms instead of trying to build them from scratch. Mastering Sklearn and similar libraries will enable you to land an ML job much more quickly, but if you don't follow through with the academic side then you'll be left implementing a bunch of pre-written algorithms that you don't quite understand. "Another good tip I got was to continue with reading some book on machine learning" Good tip indeed, you may want to find an ML book which uses the language of your choice, though if you are going to enter the fields of ML and data science Python and R are outright expected of you. One suggestion I can make is _Machine Learning for Hackers by Drew Conway and John Myles White_ which uses R, if you want Python there's _Programming Collective Intelligence by Toby Segaran_ , It'll help you get started but I've heard that the second book has become outdated because it uses outdated API's for pulling data from various sites. "picking a problem there that seemed fun" Almost any problem becomes fun with ML but the real challenge, which is constant across all problems, is the pains of acquiring data, labelling data, formatting data, normalizing data, etc so that your algorithm can properly parse it. "create a machine learning algorithm" Wouldn't recommend you start implementing your own ML algorithms at this early stage, instead get to know the existing ML techniques and understand how they work, take them apart and see what each part, each variable does. Pretty much all problems you'll encounter in the real world will fall into an already defined category such as Clustering, Labelling etc, each having its own set of ML algorithms that best suite the occasion. (Here is a nifty little chart which can help you pick an appropriate algorithm: goo.gl/yMKQt6). These algorithms are highly optimized and quick, much better than anything we could implement manually. However, if you do encounter a problem where the existing algorithms fail to deliver a satisfactory result then you can look into forming your own ML algorithms, but this is more or less heavily research oriented. Hope this long post was of some help to you and good luck :)

this is the best comment I have seen till now.

Thanks. Much appreciated.

Thanks for heads up. No doubt Andrew's course stands out of all ML video lectures but the perspective of learning you have shown is what matters. I am looking forward to Google and Sendex lectures. Thanks.

Yes

if it wasnt worldwide 3 months ago, it is worldwide now.....

NEERAJ NEGI is

Thanks man

THANK YOU! You saved me almost 30 mins.

Probably not the wisest thing to say in a Stanford video's comment section lol

@@rustin3255 i mean most cs people's dream schoolis mit its prob not that bad. if they had said harvard or UChicago it would be different

i started

everyone starts somewhere

thanks

Nicolas Bouliane ikr

is it me or does he looks like andrew yang?

@@long8398 Andrew Ng

If he is not good guy, maybe he was into his start up mot teaching others. He loves his job,it is obvious he is enjoying his current job

Koi nhi bhai

Naya wala dekhle CS229 ke naam se

Thank you!

Thanks a lot

u saved half hr of my lyf. Thanks

Dawei LIU thank you so much

No. who is he ?

@@toddmoore112 may be the most important human being in the field of machine learning and ai

@@hty96 Do you think he might be able to reprogram himself to speak more clearly?

He is Andrew Ng. You can look him up on the internet.

@@talk2thoran lmao

+Rodrigo Pinto I don't think so ....

Pada k Kembalikan ke yuotub

Definitely this time i am with you.

Vrutin Tarunchandra thank you so much

hahahahhaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa

? dont get it

The subtitle system changed. Before it was something very funny. I forget what it was now.

it's still funny now, "the checklist program plays checkers"

what's funny is that the subtitle system is also some implementation of machine learning