Is Andrew Ng’s machine learning course too mathematical?

[@Simon-kc4ml]

I think taking his mathematics for DS/ML course before taking the ML course will be really helpful

[@shafqat2379]

Bro can u suggest me as a beginner how I may start learning AI and ML

[@dannggg]

Where is this because I want to take it! Is it linear algebra and probability stuff?

[@SpeaksYourWord]

@@dannggg Coursera I believe it's free but if not apply for financial aid you'll receive it in a week or two

[@Beny123]

Guys roll your sleeves up and learn the math. It is vital to understand what happens behind all those library of code .

[@Legend18_02]

How much maths? Should I keep a tutor I am very bad at maths and also forgot what I studied in my Higher class 😅

[@Beny123]

@@Legend18_02maths is cumulative. If you have had any gaps in your knowledge go back and fill the gaps. Probability and stat , linear algebra and calculus are prerequisites for any machine learning course . You don’t have to be a master but you shouldn’t struggle when those concepts are applied to the algorithms . As per having a tutor I would say try to do it on your own first then if you find yourself spending too long and making no progress hire a tutor . All the best !

[@jaeen7665]

Agreed, once you start writing layers, you'll need to know how those matrices interact.

[@ParagPandit]

I have a doctoral degree in advanced mathematics, and yet.......... I did not understand your handwriting 😅😅

[@OfficialMazLi]

My math sucks and i understood every math related thing, I actually liked the math part more lol

[@vindolanda6974]

I did his Coursera course a few years ago. As an intro for applied ML in industry its not much use. I think its better to start with some other intro courses and get an overview of different algorithms and how they are used, strengths and weaknesses, how to implement in Python, then develop a firmer intuition using books etc, then start doing hands-on, learn more tools relevant to your work, and progressively start diving into the depths of the maths etc over time. If you start purely with maths it will take you longer to be able to work in any applied setting, months and month at least. Most practical challenges in ML are with data prep, problem definition, and process related, not the depths of an algorithm.

[@believer8754]

Yes agree atlast u will implement it in programming language so why not the sane maths covered in coding concepts

[@krox477]

If you hate maths then machine learning is not for you

[@whiteou7.]

so any recommendations of such courses?

[@iitiansouvik_07]

I'm doing a master's in mathematics from an IIT, and I found this course very interesting. I have always wanted to know the maths behind machine learning and deep learning, and this course is exactly what I was looking for.

[@GAURAVRAUL95]

You have a PhD in Mathematics 🙏🙌

[@leojack1225]

If a math PhD find hard AIdeeplearning course of NG on coursera... They did not do a math PhD. It is at most first semester of the second year in a bachelor degree.

[@user-ov4sc3fc6b]

​@leojack1225 I totally agree. There is NO way in hell that she's even finished a bachelor in STEM (especially math)

[@ArtOfTheProblem]

I try to do the intuition in my videos , keep it up

[@4thlord51]

Probably the best course to learn Machine Learning. It's pretty extensive but Andrew breaks it down really nicely.

[@rohxn16]

I would say get started with IBMs python for machine learning course, it's hand on, minimal mathematics (but its there) and a lot of implementations.

[@elegantmath9414]

good recomendation, im gonna do this course to stop working in atc's gg

[@rohxn16]

@@elegantmath9414 best of luck man

[@prasadsawant7]

much needed review of course

[@generichuman_]

Pytorch handles all the gradients and backprop stuff when you create the forward pass, but it's still good to understand how it works. A simple "hello world" example would be to implement a 5 neuron feed forward network (2 input, 2 hidden, 1 output) to learn the XOR gate. Don't use any libraries except maybe numpy (if you're in python) and implement the back prop yourself. My philosophy is usually to implement things from scratch the first time to get an intuition on how they work, then start using the established libraries.

[@siddharthpandey8516]

Agreed. This is my current goal. I'm doing the math for machine learning and ai specialization and that makes things really simple! Highly recommend it

[@771aryan]

I am on the last videos of this course. And this is very interesting and thorough course for beginning in Deep Learning and Neural Networks

[@MathForML-yd2iu]

I also have PhD in Mathematics, in the area of Mathematical Analysis. I was working on Harmonic Function Theory. What exactly was your research area?

[@pranav_c7]

I am on same journey...cheers to us🎉

[@ParulGoyal-sd1ci]

Good work

[@bikers_path]

Yes, it is too mathematical for someone seeking to build products with AI rather than actually build the underlying AI. The issue is most coaches/mentors confuse both, and sometimes they set the bar at "understand everything or don't do anything" which is extremely unfair to learners and developers. Basically anyone not seeking a career in model building/training (whatever you'd call that, AI engineer, ML engineer, Data scientist etc.) would not quite benefit from it, as much as say a course on integrating models from Hugging Face or something along those lines.

[@mimo-wx9mc]

what can you say about datacamp?

[@sagnikbiswas3268]

The math is what happens “behind the scenes.” Coming up with it on your own and deriving similar situations definitely demands great knowledge. But just using AI as a black box, much of the math can be ignored

[@tzaidi2349]

What would u say is the prerequisite knowledge I need to get the most out of this course? I waste alot of my free time learning math for fun (esp LinA) but I know next to nothing about programming.

[@rish77778]

i would say knowing linear algebra is necessary if you’re serious about learning about neural networks

[@tzaidi2349]

@@rish77778 So Ive heard! Started watching the course, pleasantly surprised by how understandable it is (so far).

[@HabibuMukhandi]

For python programming, I would highly recommend Corey Schafer's youtube channel.

[@thevishvammoliya]

I think PhD in maths doesn't matter, people with high school level math can also take the course and still understand what they need to... Andrew Ng did a great job at presenting and explaining those complex level math concepts with ease.

[@vinayakchuni1]

Lol you have a phd in math and you find Andrews course hard? You gotta be kidding 😂

[@blue5887]

that's what I thought 😭 in India if you don't know this basic stuff you can't even pass 12th imo she was being humble about the phd

[@freshprincekd]

I just finished week 1 and my math is High school level but labs made it a lot easier to understand it. The last lab took a little longer to go through on my part.

[@LevTsarin]

Is there a link for this course? That would be nice, thanks!)

[@Parcool-nw9sq]

“Day 2, I have a phd in math, I’m struggling with math”😂

[@siddharthpandey8516]

I love the math. Hate notations though 🤣 they make my head spin

[@matattz]

Yeah I wasn’t good at math in my bachelor but I have no problems at all following Ng or any DS topics because I learned the basics at my Uni. If a MATH PhD says it’s quite math heavy I have my doubts if you’re telling the truth here, do you even studied? I have massive respect for maths people

[@sso1328]

I have a maths phd and i found it hard *proceeds to write down the formulation of dot product*

[@Bigbob677]

Highly recommend the course for anyone considering taking it.

[@safashaikh2894]

Which pen are you using

[@Dhirajkumar-ls1ws]

PHD maths 😮

[@robinwang6399]

I was planning to try something like that with my bachelors in physics 😂, should I give up?

[@pfever]

Do you really have a PhD in Mathematics? I don't consider myself good in my math and I think that Andrew's Ng's course isn't particularly challenging on the math side, specially compared to other machine learning video lectures you can find online from universities.

[@samedy00]

Well, not all PhDs imply actual math knowledge. Especially if it's PhD owned by a cute girl:)😏

[@classicemmaeasy2292]

The math is what is kind of chasing me

[@mubaraq8853]

Everyone would do it if it was easy

[@dannggg]

I’m rusty with linear algebra. Does this course actually help u with some of the math stuff or you got to review first before you take the course

[@tzaidi2349]

3blue1brown has a nice LinA series on youtube. You could also watch MIT open coursewear (Gilbert Strang) but itll take a while.

[@yohanespradono5224]

so where do we start as beginner?

[@irabucc469]

Who did you slept with to get that PHD? 😂

[@bald_agent_smith]

I’ve found it actually less mathematical than I expected. This is ML and the essence of it is math.

[@TheCJD89]

That's a very good course. There is of course some math included but it's not overly complicated in my opinion. And I don't think it gets in the way of progressing through the course

[@yellowmathboard]

Great Mithuna!! yes its a great course!

[@DhirenMaji-hk5ms]

Don't demotivate

[@shivanshsinghrajput8649]

Can a beginner do this course who want to learn machine learning

[@Dom-zy1qy]

Dont be intimidated by the math guys, its really not too crazy. If you know HS calculus and matrix operations you should be able to start studying the relevant maths decent. 1 year ago i was legitimately struggling with high school trigonometry and after discovering how powerful math is, its now one of my main hobbies. I actually ended up implementing a transformer language model trained solely on text data from a discord server i own with my friends. There was about 200k messages, after training the model and setting it up to send messages via a discord bot, it started imitating people really well. Even going so far as to use custom emojis we have uploaded to our server, @ people and invite them to play games. Cool stuff like this is possible because of math. Have fun with it!

[@BenQotsa]

Is this course worth it?

[@jaeen7665]

I did this course AND a graduate level deep learning course at a pretty legitimate university. Andrew's course really only scratches the surface on DL math, it's an intro. We had to do this and more by hand, from scratch on the midterm and final exams. I disagree the math understanding isn't necessary. You have to understand matrix multiplication when writing your code and you have to understand the shape of your matrix, coz if that don't match, you'll throw an error.

[@mimo-wx9mc]

What do you recommend for people who want to deepen their skills in DL and ML?

[@pi5549]

I'd work off of Karpathy's miniseries

[@ahmedelgabry2780]

a7a phd we challenging yadeny

[@mo7amed7amdi81]

ههههههههههههههههه

[@abdelrahmanatiaa4476]

send me the link pls

[@shlokkumar6257]

The worst thing about those courses are that they make us understand mathematical concepts but not the code for that particular thing or not even tell us the libraries and the frameworks which we have to use by getting some hands on experience.

[@tiesetsomatsipa5402]

I find Andrew Ng's way teaching by explaining and giving intuitions making it easier for me to understand and also going in depth help understand faster. I tried learning from others but it's hard to understand them because they weren't diving into details but abstracted things. They explain like their talking they're talking to their colleagues, rather than to a student who has no clue and is a beginner.

[@gorkemgumusboga3717]

Actually, this is the basic version of mathematics. Check manifold deep learning.

[@adhamtito4488]

but he no give how to execute that by coding . why ? if you know any course can help me to that give me to link pleas

[@deepak_shinde]

PHD ..Great

[@KaranShah731]

Reinforced learning on human instead of machine 😂

[@Dhirajkumar-ls1ws]

You need visualization help - need to understand how these linear algebra operations happening on higher dimensions. I was able to understand transformers working by this way.

[@mohdmohsinqureshi1886]

Doing the same.

[@sunnychauhan9636]

more maths, more good with coding. even if it data science or software engineering

[@makemeinvisible1]

Andrew‘s courses are the best! - nobody else could explain the math behind it better than him. Is didactic skills are awesome.

[@GSVian8362]

Hi I am shubhashish from India I want from you to give me some tips about data science and machine learning please

[@gmog7857]

Its a useless math, no one cares

[@facts9144]

I only have further A level maths and it was relatively easy to understand, definitely worth doing the whole specialization. I give it a lot of credit for helping me study computer science at uni

[@satyabratsrikumar]

The homework assignments are just way too simple

[@T3NS0R]

I started ML with his CS229 (that blackboard one), I got saturated in 9th video out of 20 🫠