How I’d Learn Machine Learning (If I Could Start Over)

In this video, I discuss how I would learn Machine Learning in 2024 if I could start scratch again!
How to learn Python 💻: • How I'd Learn Python F...
Video Transcript: medium.com/towards-data-scien...
🍿 WATCH NEXT
• How I Became A Data Sc...
• 5 Passive Income Ideas...
• How I Keep Up-to-Date ...
🔗 LINKS MENTIONED
Linear Algebra Course • Linear Algebra - Full ...
Calculus Course • Calculus 1 - Full Coll...
Statistics Course • Statistics - A Full Un...
Khan Academy medium.com/r/?url=https%3A%2F...
Brilliant brilliant.org
Python Tutorials Point www.tutorialspoint.com/python...
Python W3 Schools www.w3schools.com/python/
Python Course • Learn Python - Full Co...
NumPy Course • Python NumPy Tutorial ...
Pandas Course • Pandas & Python for Da...
Matplotlib Course • Pandas & Python for Da...
Anaconda Course • Python for Data Scienc...
Machine Learning Specialization www.coursera.org/specializati...
Deep Learning Specialization www.coursera.org/specializati...
Kaggle www.kaggle.com
ML Algorithms From Scratch github.com/egorhowell/ML-Algo...
ML Papers of the Week github.com/dair-ai/ML-Papers-...
💌 Weekly data science tips: newsletter.egorhowell.com
MY GEAR
📷 Camera - amzn.to/3U5SLbZ
🔎 Lens - amzn.to/49sjJiL
🎤 Mic - amzn.to/49l0l76
💡 Lighting - amzn.to/43PNtod
MY TOOLS
🎵 Epidemic Sound - Where I get my music: share.epidemicsound.com/v015na
🚀 VidIQ - How I optimise my KZbin videos: vidiq.com/egorhowell
🗓️ Notion - How I manage all my stuff - affiliate.notion.so/egorhowell
(PS: Some of the links above are affiliate links that I get a kickback from 😜)
🔗 CONNECT WITH ME
Blog - / egorhowell
Twitter (X) - / egorhowell
Instagram - / egorhowell
LinkedIn - / egorhowell
⏰ TIMESTAMPS
0:00 Intro
0:25 Maths
1:59 Python
2:56 ML Libraries
4:36 ML Theory & Algorithms
5:59 Practice
7:07 How To Stand Out
8:32 Recap
WHO AM I?
Hi, I'm Egor! I am a Data Scientist with a master's in Physics currently living in London. I share data science tutorials, advice and general tech topics!
DISCLAIMERS
This content is for educational and entertainment purposes only and should not be considered as professional advice. Views and opinions are my own and do not represent or reflect the opinions of my current or past employer or any organisations I am associated with.

[@egorhowell]

Video Transcript: medium.com/towards-data-science/how-id-learn-machine-learning-if-i-could-start-over-c68d697e6a8a?sk=6eaa4385aa8852e98aca779cf5709b04

[@__Dev-iu1pn]

Data science fascinates me, but there's just so much information out there, it's hard to know where to begin and what path to take. A structured roadmap like yours would be a game-changer, providing a clear direction and highlighting key milestones for learning. Plus, I'm sure countless others in the community feel the same way.

[@egorhowell]

Thanks! I am planning on releasing a super detailed roadmap in the feature.

[@franklimmaciel]

Thanks for the tips and the suggestions courses.

[@egorhowell]

no problem, glad you liked it!!!

[@googleuser1937]

Thank you so much for the great video! I really like the way you presented it --- clean, concise, never trying to drag out the video on purpose. You're one of the few that gets straight to the point, and does not waste time, and keep things real. I look forward to learning more about the field of ML and data science!

[@egorhowell]

Thank you!!

[@Yaya2Snipey]

This is a gem of an account, deserves at least a million subscribers already ❤

[@egorhowell]

Thank you for the kind words!

[@michaelmaharlika8569]

So Awesome! Thank you for taking your time out to share this outline with so many links on ML. I am looking forward in using your suggestions to start my training asap. I also look forward to seeing more of your learning videos. Thank you gain so much Sir Egor Howell.

[@egorhowell]

Glad it was helpful!

[@aiueo8962]

Thank! Very clear in here..

[@egorhowell]

Thanks!

[@jorgecastillo-od9ri]

great video!! I can't wait to start this ML journey

[@egorhowell]

Glad you liked it!!

[@rodrix0p]

Thank you for this video!!

[@egorhowell]

You are so welcome!

[@chiragvaghela07]

Amazing! 😊

[@egorhowell]

Thank you!

[@mayanksharma6622]

Good video, you deserve more subscribers man

[@egorhowell]

Cheers mate, appreciate it!

[@ashikpirmohamed6754]

Great Video. I work with this guy. He is awesome!

[@egorhowell]

Thank you!!

[@kumarshubh9786]

The Best Roadmap on internet till now❤❤

[@egorhowell]

thank you!!

[@BradleyAidanJohnson]

Amazing content. Had some questions if you have the time. I did math at A level for a C many many years ago. I am mathematically minded but absolutely nowhere in terms of the list of mathematical concepts you listed. My original plan was to do the exact videos you posted but given my age (38)/situation (wife and kid) committing to the first math vid on linear algebra for what seems like a semester of college is a huge investment when there is no qualification at the end. We did some research and found coursera courseMathematics for Machine Learning and Data Science Specialization This offers a certificate at the end and seems to take considerably less time to learn a similar curriculum Further to this the course you recommended by Andrew Ng is now stressing that the most recent revision is aimed at people with next to no mathematical knowledge. I was wondering what your thoughts were on skipping the math element at the start to get immediately stuck into the Andrew Ng Machine Learning Specialization course given I am beyond beginner stage of python? Any reply would be appreciated. Thanks again for the info so far. I set up my blog today citing your 2 videos

[@egorhowell]

I am not against skipping the maths if you think you can make quicker progress by simply taking the course. It's all up to the individual and what's best for them. In an ideal world, you learn each thing step by step as that helps create a thorough understanding. The maths in the course is not super difficult but contains a lot of calculus (differentiation, partial derivatives etc.). So, if you are comfortable with that, then you should be fine! Let me know anymore questions. PS, link your blog I would like to check it out!

[@ahmedsadek3280]

Very informative I am also from chemistry physics background and your videos clarify the vision for me

[@egorhowell]

Thank you!!

[@DK-ox7ze]

I am currently doing Andrew NG's ML specialization course and it's great. However you mentioned that it's only entry level stuff. So just wondering what's mid level or advanced level ML and where can I get a glimpse of it? And same for Andrew's Deep learning specialization.

[@aravindnarayanan5664]

Are you talking about the course on Coursera?

[@egorhowell]

Hey! Yeah it's entry-level in a sense that it's targeted for beginners and not experts. Although, some of the things it teaches you is argubaly mid-level. In my opinion, that course will be sufficient up until the mid-level. After that, it's ideal to probably specialise in a domain. Off the top of my head, there are no courses that are advanced in all areas. However, say you want to really dig into reinforcement learning. Then, learn the basics and then try to implement recent paper. This is where the advanced knowledge will come from. Hope this helps!

[@DK-ox7ze]

@@egorhowell Thanks. Just curious as to how far are the top AI researchers of the world from someone who has done these courses, specialized a bit, implemented a research paper or two, and has some industry experience?

[@egorhowell]

@@DK-ox7ze Hmm very good question, but this is very hard to answer.I would say most researchers are quite far ahead, in their respective area, to the ones taught in the courses. This is just through the nature of their work and being very close to the cutting edge. Naturally, industry and research are different in how they approach problems.In terms of pure algorithm and theoretical knowledge it would go. 1. Courses 2. Industry 3. Researcher But that is just my opinion and it really comes down to the work that you do daily :)

[@DK-ox7ze]

@@egorhowell Based on the products I see, I think most of the cutting edge work happens in the industry rather than in academia (pure research) when it comes to AI.

[@BradleyAidanJohnson]

I've left a comment a few times now and it seems to keep getting deleted?

[@egorhowell]

just replied :)