Maths Primer
Last did maths at school? Rebuild functions, vectors, matrices, slopes, gradients and probability from pictures and small numbers, exactly as far as machine learning needs them.
- Lessons
- 7
- Study time
- 3.5 h
- Access
- All lessons free
Already comfortable with algebra, vectors, derivatives and basic probability? Skip ahead to Track 01.
Syllabus
- 0.1
Functions and graphs: a model is a function with knobs
Inputs go in, outputs come out: plot lines, parabolas and curves, turn their knobs, and see why every machine learning model is a function whose parameters get tuned.
25 minFreeLive 3D lab and Python challengebeginner- Read function notation f(x) and evaluate a function at a point
- Connect a formula to the shape of its graph, and predict how slope and intercept move a line
- See a model as a function with adjustable parameters fitted to data
- 0.2
Vectors: arrows, adding, scaling and the dot product
Drag arrows around a plane, add and scale them, measure their length, and discover the dot product as a score of how much two directions agree.
30 minFreeLive 3D lab and Python challengebeginner- Picture a vector as an arrow and as a list of numbers
- Add, subtract and scale vectors, and compute their length
- Compute a dot product and read it as alignment between two directions
- 0.3
Matrices: tables of numbers that move space
Watch a grid stretch, rotate and shear as you edit a 2×2 matrix, then see that matrix-vector multiplication is that movement applied to one arrow.
30 minFreeLive 3D lab and Python challengebeginner- Read a matrix as a transformation that moves every point of the plane
- Multiply a matrix by a vector, column by column
- Compose transformations by multiplying matrices, and see why order matters
- 0.4
Derivatives: slopes, tangents and the chain rule
Zoom into a curve until it looks straight: that slope is the derivative. Then chain functions together and watch rates of change multiply.
30 minFreeLive 3D lab and Python challengebeginner- Read a derivative as the slope of the tangent line at a point
- Estimate a derivative numerically with a small step
- Apply the power, sum and chain rules to everyday functions
- 0.5
Partial derivatives and gradients: walking downhill
Stand on a hilly surface, take the slope in each direction separately, and combine them into the gradient: the arrow that points straight uphill.
30 minFreeLive 3D lab and Python challengebeginner- Compute partial derivatives by holding every other input fixed
- Assemble the gradient vector and read it as the steepest uphill direction
- Take repeated small steps against the gradient to reach a minimum
- 0.6
Probability: chance, averages, spread and Bayes
Roll dice, draw samples and watch histograms settle: probability, the mean, the variance, the bell curve, and how Bayes' rule updates a belief with evidence.
35 minFreeLive 3D lab and Python challengebeginner- Read a probability as a long-run frequency and a histogram as a distribution whose bars add up to 1
- Compute the mean, variance and standard deviation of data and of a distribution
- Recognise the bell curve that appears when random quantities are added, and what its two parameters control
- 0.7
Exponentials, logarithms, sigmoid and softmax
Growth that multiplies, logarithms that undo it, and the two squashing functions every classifier ends with: sigmoid for yes or no, softmax for many choices.
30 minFreeLive 3D lab and Python challengebeginner- Use exponentials and logarithms and their rules with confidence
- Turn any number into a probability with the sigmoid function
- Turn a list of scores into a probability distribution with softmax