Lecture 1 | Machine Learning (Stanford)

[gauravbasu98]

i was wondering are the robots self aware??

[queerriot]

If you liked this, see stanfords online course at ml - class dot org

[callmeweezy]

I stopped watching when he said infinite dimensional space. O_o

[cmwslw]

Chill out with the 'um' complaints. English isn't his first language and he's probably a lot smarter than you. Thanks for the free info, Stanford.

[ggonlinemusic]

@cmwslw Native English speakers don't use "um"?!? If one truly listens how he enunciate each word, one would know he actually speaks "English"!

[GeekProdigyGuy]

If I had a dollar for every idiot who commented about "um" who are apparently incapable of realizing that this man is paid to pass on useful knowledge, not make speeches, then I would have more money than the sum of all those idiots' actual wealth in real life.

[Alternativerealm]

Um!

[jacf1020]

Thanks for the lecture Standford. I wish you guys could get rid of the chalk boards

[KyouJinSaMa]

if i got a dollar for every "umm" he said i would've kicked Bill Gates off the charts~~

[kagebunshin088]

Great lecture. But i noticed this was uploaded in 2008. are there any videos for the latest course taught in stanford?

[ShNaYkHs]

Great lecture, well explained, I enjoyed it.

[thebigfootme]

very technical..i didnt understand some of it..but this is a great lecture!

[sherifportrait]

142109731521074241 umm :/

[wuyanchu]

thx and god bless.

a sincere blessing from hong kong.

^_^

[Leonid05]

@lightandbeautiful what is the site? you didndt leave the link

[shrit110]

watch after 32:00

[ookete]

I enjoyed this lecture.

[nameczanin]

14 people are jealous professors from other universities.

[a1a2a3skurr1l]

From wikipedia: Big O notation (10:58) is way of defining limits (usually upper i guess) of functions.

Haha, just didn't know the name.

[clutchedhands789]

Doesn't machine learning sounds weird, because to me it sounds like we humans learning from machines! or we learning about machines!

[TheBigCollapse]

The question is how to log it? Perhaps by using xyz grid, and making the number lines logarithmic? That's my guess.

[keinplans]

@TheBigCollapse the question is how to rotate my dick 360°. maybe by moving my hips back and forwards. could work!

[PaXx]

which one of these videos involves neuro networks?

[gelliravikumar018]

The prof. starts the definition of Machine Learning at 0:32:40 with the computers' scientist Arthur Samuel.

[WSWarthog]

@gelliravikumar018

Wish your comment was at the top.

[aesnxs]

holy shit the microphone thing blew my mind!

[climber767]

Andrew Ng could you take a course for speaking more clearly??

[linagee]

In the intro where he is talking about how learning algorithms are used, he forgot to mention the hundreds of times per day various government entities use learning algorithms against you to guess if you are a terrorist.

[dylanhoggyt]

@dunstantom but you could put "completed Stanford ml-class.org course" on your resume! :P

[toothasaurus]

I feel like this 2008 lecture is already outdated.

[shebotnov]

Why would anyone dislike it? Idiots probably didnt understand it

[gespilk]

35:05 The 1959 definition is more truthful to the phrase "Machine Learning"! Mitchell's definition fails to say that the true machine learning is when P and T are defined by the machine itself!

I assert that E->(P and T)->E is the cycle the real learning machine should be doing. Out task as humans is to help the machine to learn how to set/form its P&T based on E.

All we need to do is "prime the pump" and unleash the learning machine into the world. So, this course should really be about that!

[rubelcool]

thanks alot for the knowledge that i've always been wanting to learn.

[wallghing]

13 machines didnt learn

[icommand]

I am so grateful to Professor Andrew and Stanford Uni for this video..I will surely try my best to do some research at Stanford next year.

[duppymaker]

' Algo-wiv-um ' haha

[abramswee]

thanks, andrew. due to being asperger's disorder, i can never get the proper grades to attend a proper university in my home land. this avenue of online education did open doors of ideas to me.

[JosePablo24]

Can one learn what students learn in this way? o.O

if there is like a video of each important class or something like that

is there? it would be the most awesome thing ever ! XD

[hasanmomen]

@seahawks78 you can see the course schedule on stanford (dot) edu (slash) class (slash) cs229 (slash) schedule (dot) html

[armanrainy]

Thanks Professor Andrew Ng. Thanks Stanford.

I can not leave my desk.

One Suggestion for Stanford: Subtitle for the videos will be beneficial for nonnative English speakers.

Sweden, Halmstad,Embedded and Intelligent Systems Student,

[diywebmastery]

@armanrainy He sounds fine to me.

[armanrainy]

@diywebmastery He sounds fine to me also. I guess people living in Sweden have enough English knowledge to understand what he says. I know some ''people'' who like to learn are not good in English. That's why ''khan academy'' intended to make videos in different languages. Free education for everyone.

[marcopolo60100]

Thanks for sharing

[azamstat]

Thanks a lot Professor Andrew Ng, thanks Stanford.

[ajaydublin]

Brilliant video ! Great work! Looking forward to see all the 20 video! Cheers

[ANIRUDHANROCKY]

nice initiative,, from stanford m loving it!!

[swiftset]

what's a good forum for asking questions about machine learning? for instance, when one should use the Frobenius and spectral norms to measure errors... Google didn't turn up any good sites.

[billyliang1988]

Could anyone attach some subtitles

[sfbayacm]

Thank you for sharing!!!

[AnanyaVilas]

Good that I skipped to 31:30 before I gave up on this video...I wud have missed so much of valuable info!

[Bilawallaxman]

thanks



[joni6346]

jjojjorge this is a precise example of your instinctual behavior to write kooky paragraph after kooky paragraph in bad english and hope that it means something to somebody...

[jjojjorge]

@joni6346 Who taught you to be typical that did not teach you to think? That is to really think. Since you do not really think, you are mainly governed by your untamed instincts; evident by your hostile attitude; where there is hostility, there is fear, where there is fear, there is insecurity, where there is insecurity, there is lack of real knowledge (Knowing how to know.) Like all of us, you are only reflecting your SELF whenever you think ABOUT something or someone.

[jjojjorge]

@joni6346 Too bad you did not write anything else; if you had done so I would have had the opportunity to refer it as a precise example of instinctual behaviour (We all have instincts, people may or may not dominate their instintcs in reference to a certain situation... And...)

[jjojjorge]

@joni6346 hehehehe

[jjojjorge]

Professor says something like: "Most of us use learning algorithms.... without even knowing it..." That was a prudent statement; of course if he had said all of us would have been totally imbecile, in any case, if you are aware, it should give you at least a syncretic perception (a minimum idea) of the great progressive minds in history; minds that are really aware of algorithm and much, much, much more.

[jjojjorge]

Continuing my comment: Yes, the machine can for example find bugs for example, but that s due to the engineering of algorithms, which of course ranges from a relative simple algorithm to a related relative complex algorithm, but again that is not learning. Only humans have the potential to learn. By the way, animals are driven by instinct, but that is another subject (Also not completely understood)

[jjojjorge]

Engineering algorithms upon algorithms; the more well placed they are, the more efficient the machine. More or less a reflection of the history of the programming and its applications; from machine-language to today's higher generation algorithms. But not learning at all, because basically learning requires at least a minimum of analysis to effectively render a related synthesis; furthermore, based on that synthesis one goes to next level of analysis which takes to its related synthesis, etc 

[Burhansyla]

Where can i get this kind of algorithm to separate the piano sound track of the rest of the song???

[darfunkelidas]

@Burhansyla I think melodyme managed to do this with a single recording (as opposed to many recordings at different levels). But their algorithm must be very secret.

[darfunkelidas]

@Burhansyla I meant Melodyne. Celemony Melodyne.

[nymphetamine001]

i hate this machine learning

i understand ZERO from this fucked up supject

[warlockhandler]

thnx Stanford university for this gift... always wanted to take up this course...

[lastwave235]

the book, the quantum mind, touches on these experiments.

[MrForrestli]

could anybody tell me where to download it?

[lionaneesh2009]

Gr8 work

[andrew8833]

Copying even code and even changing it is a form of plagiarism, all Universities in the UK strictly ban that.

You will find if your caught copying code from someone else (even a previous years work), you'll find you'd be kicked out of University and they wont allow you to come back on any other course.

[scottprovost]

If you are watching this video for a Free Open University Course, remember to take notes. Only your essays will be used to evaluate your progress we will not be doing the assignments mentioned in the lecture.

[many9662]

Oh I really like your video. Thank you for sharing.

[mastergsan]

what accent is he trying to pull off?

[zzuusss]

All I can say is... keep it coming, who cares about the mmhh or whatever!

[mahmoudem]

thanks

[juvaholics]

UMM UMM UMMM UMMMMM UMMMMM

[daftrhetoric]

Hearing an educator espouse unarticulated positivity feedback communicative teaching protocols is disheartening.

[MrDoctorTomas]

Ummm :D Man, it kills me :D

[headfonez]

this was ummm good!

[andrew8833]

@headfonez Its probably the best part of a strand of AI and nuerological learning, its incorporating that into computer science, if you dont understand about HCI, you'll find this incredibly hard to accomplish anything in machine learning (human computer interaction), why is a car sign instruction blue? Why are warnings red? Thats basic fundamentals of HCI, why is the play button the biggest button on a TV remote control? Again all fundamentals of HCI

[ayumilove]

Professor Andrew Ng "Umm" counter (starting from 31:40 to 1:08:40) 177 umms. Predicted 180 umms for that duration, kinda close :D

Anyways, I learn lots of stuff from here, but the umms kinda annoy me from lecture immersion.

[ottoreichert]

Great thanks for Stanford for the GREAT COURSE!!!

[aperezNWO]

Thanks for sharing, from Colombia

[aperezNWO]

Thanks from sharing from Colombia.

Gracias por compartir este video desde Colombia.

[amrosik]

are these lectures helpful? please give me an reply, then i can start to download them, please

[manish1golu]

thanks to the university...............

nd also to the prof.... who give his important time to the student like me...

[gianantonio100]

Professor NG rocks !

[interted]

where is Andrew from?

[interted]

what do you think of Alpaydin's book? is it a good textbook to consult in parallel with these lectures?

[wizztjh]

thanks for posting this nice video , it really help me in understanding of machine learning!

[kjsdffkjweh8]

Thanks for posting all these awesome lectures guys!

[nielsww]

Will be watching them all. Thanks for uploading these great classes!

[matthewrobertson03]

I'm wondering why I spent so much money on a university degree -_-

[dunstantom]

@matthewrobertson03 because you can't put "watched youtube" on a resume :P

[astroboomboy]

@matthewrobertson03 Do you regret having gone to uni? Wouldn't it have been hard to find a job without a degree? But I think I understand your point, we can probably learn more through the internet than at a geographically local place, maybe universities will cease to exist in the future...

[andrew8833]

@matthewrobertson03 You might think that but who is there to review your work if you did not go to University?

Unless you have someone to help you with this kind of stuff at home then fair enough.

[griesrt]

@matthewrobertson03 need that credibility indicator

[TheMr247]

Thankyou!! fantastic post!! will be watching them all - and looking out for more :)

[tleeuwenburg]

I would just like to echo another "Thank You".

[BrianVandrian]

thanks for posting!

[edyinthesky]

does he mean AI instead of "machine learning"???

[Kornmeister]

"Machine learning" is basically when a machine is able to improve it's AI on it's own.

[Compact3]

Isn't AI equal to machine learning?

[someonefromsomewere1]

@Compact3

Not exactly,

AI is just the computer following certain instructions based on predefined circumstances, but machine learning is when the machine starts to learn from its mistakes and don't make them a second time. (or something like that )

[exetion]

@someonefromsomewere1 wrong, machine learning is a part of AI

[rafaelhsouza]

No.

[notbored12]

Machine learning used to be called "statistical pattern recognition" but a more vague and impressive sounding title probably attracts more funding. AI has a slightly different meaning than machine learning but this comment is too brief to go into detail, google is your friend.

[allenst]

wow the guy has a personality for a wang

[woodengun]

but still better than yours fella!

[sg04f]

Dr Ng mentions that we can find the review classes for the prerequisites online. Where are they posted?

[sudo333]

Professor Ng is so great.

thanks.!!

[therobz98]

orm.. orm.. orm...

[disprefer]

NOTE: skip to 31:30 if you're here to learn about machine learning.

[MisappliedRhetoric]

31:30? Good God.

Thanks.

[happycamperjack]

@disprefer more like 32:55

[sabriath]

@happycamperjack : Actually it's more like 1:08:41 considering this doesn't teach you a dam thing. There were more "ums" than definitions...and more definitions and explanations than implementation. Useless, utterly useless.

[BillNyeTheScienceGuy]

@sabriath Just by looking at the title of this video, one could easily note that this is the very first lecture in a whole class about machine learning. If you want to learn a whole discipline of computer science in a 1 hour video then you are obviously looking in the wrong place. As for the use of "ums" in the lecture, I take it you yourself are a university professor and your quality of presentation is that much better than this man. In short, get off your high horse asshole.

[sabriath]

@BillNyeTheScienceGuy : The point of me making that comment was that there is hardly any research online about it, and what information I can find is all in lectures of a bunch of definitions that I could easily get off wikipedia. There is no reason to go off definitions in a public lecture, that's left for homework. When I was in college, lectures were about production, showing WHY it works and HOW it works....not what the words mean.

[TheDvloper]

@disprefer I would say 32:40

[Fuglebolle]

@disprefer Thank you

[OZZl3]

@disprefer more like skip to 32:38 - if you are here to learn about maschine learning

[outspan87]

umm... umm... :D

great lecture

[proximo20]

i hope not all examples are similar to tumor sizes and their malignancy rates.

[moekelok]

I cannot convey how grateful I am that an academic institution of the stature of Stanford is generous enough to offer fascinating information like this for free. Cheers!

[khaldoon2300]

Thanks a lot for offering the course on Youtube. I really really appreciate it.

It seems very useful and it will give me an opportunity learning something valuable for free!!

[heocon19ca]

I think his accent is fine! Not perfect but it's really fine.

[pinochet222]

it starts at 33:00 lol

[Neuromancer360]

cheers to university's offering thier courses online free as many of us cannot either afford classes or take time off work to go to school

[szuo2009]

thanks for sharing the amazing series

[7errain]

Is this course undergrad or graduate at stanford?

[Ekusuos007]

The 200-series courses at Stanford are marked as "advanced undergraduate/beginning graduate".

[leelguy]

Thanks stanford for sharing these great courses!

[ferdielicious]

tools are meant to make our life simpler. "Teaching" machines to "learn" will not give it consciousness, instead it will even make us realize more the complexity of our mind and the power of the creator who designed it. If you disagree with machine learning, then you would be seeing profane words here in you tube, no software would keep me from telling you that you're such a closed-minded ignorant!

[MarcoAdeAlmeidaSilva]

Oh no, argument by design again? What "creator" could you be talking about?

[roywwcheng]

One should not watch too many movies... :D

[Alterstateresearcher]

So many disiplines present at this lecture, smells like a singularity.

[Rombizio]

You mean disciplines...with a C?

[efittery]

Anybody care to form a study group for Machine learning?

[neoniator]

I am in for group.

[habozaid]

I am in too

[YetWhyNot]

me 3

[tr0p]

According to the course website there is no required textbook for the course, but supplementary texts are recommended:

Christopher Bishop, Pattern Recognition and Machine Learning. Springer, 2006.

Richard Duda, Peter Hart and David Stork, Pattern Classification, 2nd ed. John Wiley & Sons, 2001.

Tom Mitchell, Machine Learning. McGraw-Hill, 1997.

Richard Sutton and Andrew Barto, Reinforcement Learning: An introduction. MIT Press, 1998

[ssampanthan]

Um ... do us youtube viewers have to follow the honor code too?

[gijs14]

aniko voilations?

[kuroshe1]

but he is giving some good inormation, thank you .

[kuroshe1]

i am not very impressed with this lecture, i guess stanford is aslo over rated.....

[renatoaz]

who cares?

[NGUY0035]

thank you Standford for uploading those videos. They help me a lot for self-improve my skills and knowledge !!!

[alfiecarter111]

Brilliant

[briantonio]

It great :D but he is useing .erm. many time´s.. but its still great! :D

[lqk1985]

I can't stand his continuous sound "erm". Is there any one can use Pattern Recognition to eliminate this sound!~

[cekkong]

what a good lecture...

I dont need any paper

I just have to watch this and I understand

so good

thank u so much...

[tteallcot]

good lecturer except annoying ass whiny nose voice... damn.

only if

[6369703]

boring and hand waving stuffs

[ewonsnappea]

Thank you Professor, for sharing your lectures with the public on You Tube. Its a contemporary example of global goodwill ambassdorship! You and Stanford both are to be commended!

E