How do I start learning machine learning with a weak math background?
Dropped math after high school. Every ML course opens with linear algebra and I bounce off immediately. Is there a practical way in?
73
2 answers1.3K views2 Answers
Accepted answer
Yes — flip the order. Code first, math when it blocks you.
Start with fast.ai's 'Practical Deep Learning' course: you train a working image classifier in lesson one and learn the theory backwards from working code. Alternatively Kaggle's free micro-courses (Python → Intro to ML) get you building models in days.
Then learn math on demand: when you hit gradient descent and want to know WHY it works, watch 3Blue1Brown's calculus and linear algebra series — visual, intuitive, zero prerequisites. Khan Academy fills specific gaps.
The dirty secret: day-to-day applied ML is mostly data cleaning, feature engineering and evaluation judgment. The people deriving equations from scratch are researchers. You can be a genuinely good ML practitioner with intuition-level math plus strong Python.
15
Anecdata: I got an ML job with commerce-stream math. What got me through interviews was being able to EXPLAIN bias-variance tradeoff, overfitting, and train/test leakage in plain words with examples from my projects — not derive anything. Depth of intuition beats breadth of formulas.
7
Related questions
- How do I cope with anxiety before exams or big presentations?1 answer
- How many hours a day should I actually study?1 answer
- How do I learn UI/UX design with no art background?2 answers
- How do I stay focused while studying when my phone is right there?1 answer
- What's the difference between AI, machine learning and deep learning?1 answer
- Is data science still a good career in 2026 or is the field saturated?1 answer