# Maths for machine learning course

**URL:** <https://community.deeplearning.ai/t/maths-for-machine-learning-course/652122>\
**Category:** Supervised ML: Regression and Classification\
**Created:** [June 23, 2024, 8:57am UTC](https://community.deeplearning.ai/t/maths-for-machine-learning-course/652122 "2024-06-23T08:57:41Z")\
**Posts on this page:** 20\
**Page:** 1

<div class="post-metadata">

**Author:** ![Vidushi\_29](https://avatars.discourse-cdn.com/v4/letter/v/ea5d25/32.png) [@Vidushi\_29](https://community.deeplearning.ai/u/Vidushi_29)\
**Post date:** [June 23, 2024, 8:57am UTC](https://community.deeplearning.ai/t/maths-for-machine-learning-course/652122/1 "2024-06-23T08:57:41Z")

</div>

Do we need to learn mathematics seperately from somewhere before starting this Machine Learning course, if yes, then please suggest some interesting resources for that.

---

<div class="post-metadata">

**Author:** ![Alireza\_Saei](https://sea1.discourse-cdn.com/dlai/user_avatar/community.deeplearning.ai/alireza_saei/32/85175_2.png) [@Alireza\_Saei](https://community.deeplearning.ai/u/Alireza_Saei)\
**Post date:** [June 23, 2024, 9:30am UTC](https://community.deeplearning.ai/t/maths-for-machine-learning-course/652122/2 "2024-06-23T09:30:13Z")

</div>

Hi @Vidushi_29

It is generally recommended to have knowledge of mathematics before starting to learn Machine Learning. Knowing the following subjects can make your learning journey easier:

1. Linear Algebra
2. Calculus
3. Probabilities & Statistics

In coursera, there is a `Mathematics for Machine Learning Specialization` that you can start with. Otherwise, there are lots channels on YouTube (3Blue1Brown, Khan Academy, StanfordOnline, etc.) or books that you can start with. However, you can also start your courses right away!

Hope this help, feel free to ask if you need further assistance!

---

<div class="post-metadata">

**Author:** ![TMosh](https://sea1.discourse-cdn.com/dlai/user_avatar/community.deeplearning.ai/tmosh/32/9321_2.png) [@TMosh](https://community.deeplearning.ai/u/TMosh)\
**Post date:** [June 23, 2024, 6:44pm UTC](https://community.deeplearning.ai/t/maths-for-machine-learning-course/652122/3 "2024-06-23T18:44:35Z")

</div>

The Math for Machine Learning course covers a lot of topics that are not related to Machine Learning. It is all interesting mathematics however. Whether to attend it depends on your interests.

---

<div class="post-metadata">

**Author:** ![rmwkwok](https://sea1.discourse-cdn.com/dlai/user_avatar/community.deeplearning.ai/rmwkwok/32/126977_2.png) [@rmwkwok](https://community.deeplearning.ai/u/rmwkwok)\
**Post date:** [June 24, 2024, 12:19am UTC](https://community.deeplearning.ai/t/maths-for-machine-learning-course/652122/4 "2024-06-24T00:19:55Z")

</div>

Hello, @Vidushi_29,

I think basic maths and basic Python are enough for this Machine Learning Specialization (MLS).

Being good at algebra, calculus and probability is a bonus as they will give you more insight of ML concepts, but it actually takes time and discussions to really fully leverage the maths knowledge, so without extensive use and discussion of maths, I do not think just taking a maths course will give you 100% of that bonus.

Maths will be essential if you go deep into the ML world, but it should be considered as a longer term study instead of just a prerequisite for this MLS, not to mention that basic maths should be sufficient here, because this MLS is considered beginner level .

Cheers,  
Raymond

---

<div class="post-metadata">

**Author:** ![Vidushi\_29](https://avatars.discourse-cdn.com/v4/letter/v/ea5d25/32.png) [@Vidushi\_29](https://community.deeplearning.ai/u/Vidushi_29)\
**Post date:** [June 24, 2024, 6:01am UTC](https://community.deeplearning.ai/t/maths-for-machine-learning-course/652122/6 "2024-06-24T06:01:12Z")

</div>

Will it be fine if I just start the course of ML without learning math seperately?

---

<div class="post-metadata">

**Author:** ![Vidushi\_29](https://avatars.discourse-cdn.com/v4/letter/v/ea5d25/32.png) [@Vidushi\_29](https://community.deeplearning.ai/u/Vidushi_29)\
**Post date:** [June 24, 2024, 6:01am UTC](https://community.deeplearning.ai/t/maths-for-machine-learning-course/652122/7 "2024-06-24T06:01:54Z")

</div>

@ rmwkwok  
Thanks a lot for your advice. Can you please specifically tell what to study in maths, or should I just kickstart my journey of “Machine Learning Specialization”? That will be a great help.

---

<div class="post-metadata">

**Author:** ![Vidushi\_29](https://avatars.discourse-cdn.com/v4/letter/v/ea5d25/32.png) [@Vidushi\_29](https://community.deeplearning.ai/u/Vidushi_29)\
**Post date:** [June 24, 2024, 6:04am UTC](https://community.deeplearning.ai/t/maths-for-machine-learning-course/652122/8 "2024-06-24T06:04:13Z")

</div>

@ Alireza\_Saei  
Thanks for your advice. Should I need to learn everything in deep? I am very confused like what should I start and from where.

---

<div class="post-metadata">

**Author:** ![TMosh](https://sea1.discourse-cdn.com/dlai/user_avatar/community.deeplearning.ai/tmosh/32/9321_2.png) [@TMosh](https://community.deeplearning.ai/u/TMosh)\
**Post date:** [June 24, 2024, 6:33am UTC](https://community.deeplearning.ai/t/maths-for-machine-learning-course/652122/9 "2024-06-24T06:33:11Z")

</div>

> [@Vidushi\_29](#):
>
> Will it be fine if I just start the course of ML without learning math seperately?

It depends on how much math you already understand.

Note that the Machine Learning Specialization (MLS) existed for a long time before the “Math for Machine Learning” (M4ML) course was created. Since some students had a lot of questions about the underlying mathematical methods, the M4ML course was created for those who are curious about the details.

The MLS course assumes only that you understand basic algebra (i.e the equation of a straight line), and a little bit of statistics (like means, standard deviations, and probabilities for rolling dice).

---

<div class="post-metadata">

**Author:** ![Vidushi\_29](https://avatars.discourse-cdn.com/v4/letter/v/ea5d25/32.png) [@Vidushi\_29](https://community.deeplearning.ai/u/Vidushi_29)\
**Post date:** [June 24, 2024, 6:41am UTC](https://community.deeplearning.ai/t/maths-for-machine-learning-course/652122/10 "2024-06-24T06:41:53Z")

</div>

Oh okay. This basic math I know. Rest they will tell in the MLS course itself right?

---

<div class="post-metadata">

**Author:** ![TMosh](https://sea1.discourse-cdn.com/dlai/user_avatar/community.deeplearning.ai/tmosh/32/9321_2.png) [@TMosh](https://community.deeplearning.ai/u/TMosh)\
**Post date:** [June 24, 2024, 7:17am UTC](https://community.deeplearning.ai/t/maths-for-machine-learning-course/652122/11 "2024-06-24T07:17:21Z")

</div>

> [@Vidushi\_29](#):
>
> Rest they will tell in the MLS course itself right?

Yes.

---

<div class="post-metadata">

**Author:** ![Vidushi\_29](https://avatars.discourse-cdn.com/v4/letter/v/ea5d25/32.png) [@Vidushi\_29](https://community.deeplearning.ai/u/Vidushi_29)\
**Post date:** [June 24, 2024, 7:37am UTC](https://community.deeplearning.ai/t/maths-for-machine-learning-course/652122/12 "2024-06-24T07:37:05Z")

</div>

Thank you so much for your time.

---

<div class="post-metadata">

**Author:** ![Alireza\_Saei](https://sea1.discourse-cdn.com/dlai/user_avatar/community.deeplearning.ai/alireza_saei/32/85175_2.png) [@Alireza\_Saei](https://community.deeplearning.ai/u/Alireza_Saei)\
**Post date:** [June 24, 2024, 8:31am UTC](https://community.deeplearning.ai/t/maths-for-machine-learning-course/652122/13 "2024-06-24T08:31:33Z")

</div>

Hi again,

Not that deep! Just an understanding of each subject mentioned is enough for MLS.

---

<div class="post-metadata">

**Author:** ![Vidushi\_29](https://avatars.discourse-cdn.com/v4/letter/v/ea5d25/32.png) [@Vidushi\_29](https://community.deeplearning.ai/u/Vidushi_29)\
**Post date:** [June 24, 2024, 8:56am UTC](https://community.deeplearning.ai/t/maths-for-machine-learning-course/652122/14 "2024-06-24T08:56:52Z")

</div>

Fine. Thanks a lot!

---

<div class="post-metadata">

**Author:** ![rmwkwok](https://sea1.discourse-cdn.com/dlai/user_avatar/community.deeplearning.ai/rmwkwok/32/126977_2.png) [@rmwkwok](https://community.deeplearning.ai/u/rmwkwok)\
**Post date:** [June 24, 2024, 9:37am UTC](https://community.deeplearning.ai/t/maths-for-machine-learning-course/652122/15 "2024-06-24T09:37:42Z")

</div>

@Vidushi_29, for example, in this MLS, you will see gradient formula derived for you. What will you do? Can you choose to accept the formula instead of proving it yourself? The course will not ask you to prove it, but will you insist it yourself? Sometimes some learners do insist it for their own reasons 😉

There is nothing wrong to prove it yourself, but if you were not strong in calculus, then could you hold it, finish the MLS first, and decide whether you want to advance your calculus skills to prove it later?

This course will not ask you to do advanced maths. If you follow the flow of this specialization and leave any of your own maths adventure to a later time, then you may start MLS right the way!

---

<div class="post-metadata">

**Author:** ![CaulyChybbs](https://avatars.discourse-cdn.com/v4/letter/c/91b2a8/32.png) [@CaulyChybbs](https://community.deeplearning.ai/u/CaulyChybbs)\
**Post date:** [June 28, 2024, 3:30pm UTC](https://community.deeplearning.ai/t/maths-for-machine-learning-course/652122/16 "2024-06-28T15:30:26Z")

</div>

I have practical experience with this. My advise is to learn the concepts of maths for machine learning before diving deep into ML. You can have for some algorithms the mathematical concepts hidden behind the scene with libraries like scikitlearn doing the mathematical work, but when it comes to model explainability, you might really struggle if you have little mathematical and/or statistical background.

---

<div class="post-metadata">

**Author:** ![iamjavidanakbarov](https://avatars.discourse-cdn.com/v4/letter/i/f08c70/32.png) [@iamjavidanakbarov](https://community.deeplearning.ai/u/iamjavidanakbarov)\
**Post date:** [July 14, 2025, 3:18pm UTC](https://community.deeplearning.ai/t/maths-for-machine-learning-course/652122/17 "2025-07-14T15:18:00Z")

</div>

Well, honeslty it does. I dont know why did you come up with such opinion. Can you explain further?

---

<div class="post-metadata">

**Author:** ![TMosh](https://sea1.discourse-cdn.com/dlai/user_avatar/community.deeplearning.ai/tmosh/32/9321_2.png) [@TMosh](https://community.deeplearning.ai/u/TMosh)\
**Post date:** [July 14, 2025, 3:55pm UTC](https://community.deeplearning.ai/t/maths-for-machine-learning-course/652122/18 "2025-07-14T15:55:53Z")

</div>

Are you asking me to explain my post from one year ago?

---

<div class="post-metadata">

**Author:** ![iamjavidanakbarov](https://avatars.discourse-cdn.com/v4/letter/i/f08c70/32.png) [@iamjavidanakbarov](https://community.deeplearning.ai/u/iamjavidanakbarov)\
**Post date:** [July 14, 2025, 4:10pm UTC](https://community.deeplearning.ai/t/maths-for-machine-learning-course/652122/19 "2025-07-14T16:10:59Z")

</div>

1, 10, 100. It doesn’t really matter to anyone. As long as the opinion is there, be prepared to be asked for clarification.

---

<div class="post-metadata">

**Author:** ![TMosh](https://sea1.discourse-cdn.com/dlai/user_avatar/community.deeplearning.ai/tmosh/32/9321_2.png) [@TMosh](https://community.deeplearning.ai/u/TMosh)\
**Post date:** [July 14, 2025, 4:38pm UTC](https://community.deeplearning.ai/t/maths-for-machine-learning-course/652122/20 "2025-07-14T16:38:57Z")

</div>

You may certainly ask.

I have the option of how to reply.

---

<div class="post-metadata">

**Author:** ![mcdermott.d.r](https://sea1.discourse-cdn.com/dlai/user_avatar/community.deeplearning.ai/mcdermott.d.r/32/123573_2.png) [@mcdermott.d.r](https://community.deeplearning.ai/u/mcdermott.d.r)\
**Post date:** [January 19, 2026, 2:34am UTC](https://community.deeplearning.ai/t/maths-for-machine-learning-course/652122/21 "2026-01-19T02:34:19Z")

</div>

I’m doing the course now - I was terrible at High School maths (this was a long time ago) and I am working through this course.

It is quite hard but with enough dedication I am pushing through the courses and work. It is definitely not impossible.

Would a math refresher help? Potentially… I have a bad habit of getting distracted though so I would rather just keep working through these courses here and try to understand what I can and then pick up some additional courses elsewhere if needed.

[Next page](https://community.deeplearning.ai/t/maths-for-machine-learning-course/652122.md?page=2)
