You DONT Need Math for Machine Learning

I talk a little bit about why I think you don't need math for Machine Learning. There are certainly some cases where math is very important for machine learning, especially in theory, but when you talk about ML on the whole, you can often get great use out of the field and AI without knowing a whole lot of math. The math can be helpful, but it is not always essential for learning and implementing machine learning.

Some examples that do use math are understanding gradient descent, many sampling algorithms, performing calculations or implementing algorithms like naive bayes, SVMs, or even simply linear regression.

[estiaquzzaman]

Which platform do you recommend to learn Machine learning and deep learning as well without math???

[rssgtzk]

The situation is changed after 3 years. Now, we don't understand SOTA component or papers without deep familiar math... I'm really frustrating recently because of this.

[RohanPaul-AI]

Very well-structured explanation on this issue. Subscribed to your channel and would love to see more videos from you. Thanks !!

[vrsightseer]

Hi, I'm a python developer and would like to build a model that takes several images as inputs, and then outputs another image. The model should train until the output image matches the correct image for those inputs. It's been 30 years since we did algebra at school, and it was no where near as well covered as it is today. Every time I try and learn machine learning via udemy courses or books, it get's very math heavy quite quickly and I can't continue. Even if I watch math for machine learning tutorials, after about 8 video's I find I don't have sufficient basic algebra knowledge to understand. For what I'm trying to accomplish, do I need an understanding of algebra/linear algebra and calculus? If not, could you recommend a course that will teach me what I need without the math knowledge? Sort of a machine learning for morons who just want to build apps :D

I'm willing to learn all the math, but I just have a hunch that for what I want to do, it'll be unnecessary, but it's proving difficult to get to a sufficient skill level with machine learning to know for sure, because all of the courses I've found so far, use complex math from the get go.

[jasoncarrasquillo]

Thanks for this. Very encouraging for someone who has a hard time with math. In a 6 month AI/ML post graduate program and leaning towards computer vision. Learning PyTorch and Yolo for license plate reading with a tutor. Excited to add this new skill!

[alteshaus]

Thank you very much!

I really wanted to know how important it is to know the math. Of course the math is useful, but not necessary. Great video, i am really happy now

[emmanueldamour]

Question you’re saying you don’t need math to write brand new algorithms for AI and ML?

Because last time I check when ever you’re innovation measurements must be done how do you do that without math?

[dhyanpandya]

I was too bad at calculus but I want to pursue AI as career then what should I do?

[MarcusAureliusSeneca]

1. All the people who say you dont need math for ML, already know math
2. If I know ML and another guy knows ML just as well but also knows Maths he is gonna get the job
3. Knowing maths is proof you are ready to invest a lot of effort into really complex and boring tasks relentlessly and you kind of have paid your dues to get the ML position
4. You said we need a little calculus. You can’t afford to know just a little calculus if you dont already know all the maths taught in school on which Calculus is built upon and if you do, youre kinda already almost there anyway. If you didnt know maths like me, you would see how scary even the most fundamental things seem and how you dont know what weird symbols represent in a formula
5. You only need to encounter one simple formula in a linear regression turorial and not understand it to feel like an imposter
6. Try reading an ML paper without knowing the maths
7. Try talking to math-savvy ML engineers at work when they use mathematical formulas or terminology and see how you feel
8. There is a difference between what you do at work (maybe you can survive without the math) and what you need to do to get the job, so maybe you can get by without the math but you cant get to the job in the first place


i.e. the title of the video should be “you dont USE maths in ML” but you do need it to even be there

[ExistentialSadness]

I've got a Junior ML Developer job recently. I only have 6 months of experience in Python Backend. I have economics degrees (bachelor's and master's). I studied statistics and math during university but nothing crazy. So, you think, I can be an ML developer without insane math skills?
People say I won't build ML algos from scratch and just use TensorFlow, Keras and basically implement stuff.
Is this true?

[valentinrafael]

Math literacy, even amongst computer scientists, is needed. Otherwise you think that AI there is real. And just like that lady on the plane said, "that mf aint real".

[dank]

Very good video!

What would be your thoughts on Research Engineering? 🧐🤔

[ajaykiranchundi]

Very helpful and subscribed :) . Thank you