Indie Degree
GitHubSign in
← Programme

Mathematics for Machine Learning

AIE-201

3 credits · 70h required

Targeted, not an undergraduate sequence. The failure mode of maths-for-ML is spending a year on proofs and never arriving at the machine learning, so every module here ends in something computed by hand on an object from your own work and then checked against a library. You are not being taught to prove theorems. You are being taught to read a paper's equations without skipping them, and to know what an optimiser is actually doing when a training run goes wrong.

The lab. Your own corpora, because a matrix built from your own data is a fact rather than a textbook exercise. Indie Degree's 203 verified resources make a term-document matrix whose singular values tell you something true about how repetitive your own reading list is. Career Side Quests' scoring pipeline is a real loss function with real variance. The maths lands differently when the matrix is yours.

By the end you can

  • Manipulate matrices, eigendecompositions and projections fluently enough to read architecture papers
  • Differentiate through a computational graph by hand
  • Reason about distributions, expectation and variance in a loss function
  • Explain gradient descent and its common variants from the mathematics, not the metaphor

What this builds

  • Mathematics for machine learning · supporting0/19 demonstrated

Next assessment. Decompose your own corpus — 180 min, tier 1. Machine-checkable, so you can attempt it before reading anything and find out whether you need to.

0/19 assessed items demonstrated · 0/14 read or watched. The second number is how you close a gap; only the first is evidence that it closed.

0 of 19 demonstrated · 0/14 read

0m of 70h

M1 · Linear algebra you actually need

0/7 · 15h

  • lecture · 2h 30m · tier 0 self-marked

    Essence of linear algebra — 3Blue1Brown

  • lecture · 2h 30m · tier 0 self-marked

    Essence of linear algebra — 3Blue1Brown

  • reading · 2h · tier 0 self-marked

    Mathematics for Machine Learning

  • reading · 2h · tier 0 self-marked

    Mathematics for Machine Learning

  • assignment · 3h · tier 1 machine-verified

  • assignment · 2h 45m · tier 1 machine-verified

  • retention · 15m · tier 1 machine-verified

M2 · Differentiating a computational graph

0/7 · 14h

  • lecture · 2h · tier 0 self-marked

    Essence of calculus — 3Blue1Brown

  • lecture · 2h · tier 0 self-marked

    Essence of calculus — 3Blue1Brown

  • reading · 2h · tier 0 self-marked

    Mathematics for Machine Learning

  • assignment · 3h · tier 1 machine-verified

  • assignment · 2h 30m · tier 1 machine-verified

  • assignment · 2h 15m · tier 2 panel-assessed

  • retention · 15m · tier 1 machine-verified

M3 · Probability, and what a loss function is measuring

0/6 · 14h

M4 · Optimisation

0/6 · 14h

  • reading · 2h 30m · tier 0 self-marked

    Mathematics for Machine Learning

  • reading · 2h 30m · tier 0 self-marked

    Deep Learning

  • assignment · 3h · tier 1 machine-verified

  • assignment · 2h 45m · tier 2 panel-assessed

  • assignment · 3h · tier 1 machine-verified

  • retention · 15m · tier 1 machine-verified

M5 · The mathematics of something you already use

0/7 · 13h

  • lecture · 1h 15m · tier 0 self-marked

    Neural networks — 3Blue1Brown

  • reading · 2h · tier 0 self-marked

    Attention Is All You Need

  • assignment · 2h 30m · tier 2 panel-assessed

  • project · 3h · tier 2 panel-assessed

  • project · 3h · tier 3 artifact

  • defense · 1h · tier 4 defended

  • retention · 15m · tier 1 machine-verified