How to actually learn mathematics
Mathematics has the widest gap between watching and doing of any subject: a lecture can feel perfectly clear while the skill it describes remains entirely absent. The method is unglamorous and works — worked examples, then unseen problems, spaced and mixed.
The method
Work fully-shown examples, then unseen problems, until the method survives without the example in front of you.
Problem sets with rising difficulty — and a worked solution for every problem, including the ones you got right.
Where it goes wrong
A whole answer marked wrong for one early slip.
Separate the two failures: method choice and execution. A worked solution for every problem — including ones you got right — shows which one actually failed.
A prerequisite from years ago silently blocking the new topic.
This is the most common invisible wall in mathematics. A placement that finds your real starting point beats any amount of effort applied at the wrong level.
What mastery actually looks like
Not a grade — observable abilities. A plan should be able to say which of these it has evidence for:
- Accuracy — gets the right answer when the method is known.
- Method choice — picks the right approach for an unseen problem.
- Reasoning — shows a chain of steps that would convince a marker.
- Fluency — works standard problems fast enough for an exam.
Keeping it
Spaced problem sets that mix old topics in. The method decays fast without use.
Common questions
Why do I understand the lecture but fail the problems?
Understanding an explanation and executing a method are different skills. The engine plans maths at 60% problem-solving for exactly this reason — the lecture is a quarter of the work.
Should I review problems I got right?
Yes — read the worked solution anyway. A right answer with a fragile method fails on the exam variant; the solution shows the method the marker expects.
How do I stop forgetting earlier topics?
Interleave: problem sets that mix old topics into new ones. Blocked practice feels better and retains worse.
Engine values reviewed 2026-09-01. Related: data science, accounting, statistics.
Akiri turns this into your actual week: tell it your goal and your date, and it measures where you're starting from, builds the plan backwards from the deadline with this discipline's own rhythm, and rebuilds it as you go — in English, French, Spanish, Hindi or Chinese.