Index Ai Academy
Foundation · Shapes, angles and waves

Waves: height, speed and shift

Amplitude, period and phase, sampling a wave, and position codes in transformers.

About 37 minutes · Free lesson

What it covers6

Why it matters

Anything that repeats, a sound, a season, a heartbeat, can be described as a wave. A sine wave is simply the height of the turning arm from the last lesson, drawn out over time. Three numbers set its shape: how tall, how fast and how far shifted. In AI, sound reaches a model as a long list of wave samples, and transformers use waves of different speeds to tell the first word of a sentence from the fiftieth. This lesson reads the three dials off a formula, samples a wave into a list the way a microphone does, and builds the position codes a transformer uses.

What you'll be able to do

  • Read a sine wave as the height of a turning arm over time
  • Name a wave's amplitude, period and phase shift, and read them from y = A sin(kx + φ)
  • Sample a wave into a list in Python, and see what too few samples lose
  • See sound as a list of samples, and how a transformer tells positions apart with waves
Quick check
Have a guess before you start. An arm turns round and round. If you plot its height over time, what shape do you get?
  1. A wave that rises and falls forever
  2. A circle drawn round and round
  3. A single bump that then stays flat
  4. A straight line rising steadily upward
Show the answer

A wave that rises and falls forever

The height goes up to the top, down through the middle to the bottom, and back, every turn: a repeating wave. That wave is the sine wave.

The math: a turning arm, drawn over time

Purpose. Describe anything that repeats.

In the last lesson, an arm of length 1 at angle x had height sin x. Let the angle grow steadily, like time passing, and plot the height: it rises from 0 to 1 at a quarter turn, falls back to 0 at a half turn, down to −1 at three quarters, and back to 0 after a full turn. Then it all repeats. That's the sine wave.

Three numbers stretch and slide it into any wave you need:

y=Asin⁡(kx+φ)y = A \sin(kx + \varphi)

In plain words: A sets how tall, k how fast it repeats, and φ (phi) how far it's slid along.

Quick check
From the last lesson: what is sin 90°?
  1. 0.5
  2. 0
  3. −1
  4. 1
Show the answer

1

At 90° the arm points straight up, so its height is 1, the top of the wave.

Sampling. A computer can't store a smooth curve, so it stores the wave's height at evenly spaced points: samples. Watch a wave with amplitude 2 sampled 8 times across one full cycle. In the picture, an arm 2 long turns an eighth of a turn each step, and its tip's height becomes the next point of the wave beside it. Straight lines join the points, because the samples are all the computer keeps.

Try it now

Then set the samples to 2 and press Use these numbers: both samples land on 0, and the joined wave in the picture is flat. Try 4, then 16.

Symbols and formula. The three dials:

amplitude=Aperiod=2πkshift=φ\text{amplitude} = A \qquad \text{period} = \frac{2\pi}{k} \qquad \text{shift} = \varphi

In plain words: the amplitude is the height of a peak. The period is how long one full cycle takes: a bigger k packs the cycles closer. The phase shift slides the wave along without changing its shape.

The three pictures below each turn one dial. Each starts from y = sin x, drawn dashed, and draws the new wave thick on top, so you can see exactly what that one number changed.

So y = 3 sin(2x) has peaks of 3 and repeats every 2π/2 = π, twice as often as plain sin x. Shifting sin x by φ = π/2 makes it start at its peak instead of at 0: that's exactly cos x, the same wave a quarter turn ahead.

Reading a wave from its formula. Take y = 0.5 sin(4x). The number in front, 0.5, is the amplitude, so the wave swings between −0.5 and 0.5. The number multiplying x, 4, is the speed, so the period is 2π/4 = π/2 ≈ 1.57, and four cycles fit where sin x makes one. There's nothing added inside the brackets, so there's no shift: the wave starts at 0 and rises. Reading all three dials first tells you what to expect before you plot a single point, and a plot that disagrees points straight at the slip. A minus sign in front, as in −2 sin x, flips the wave upside down: it starts at 0 and falls first, with peaks still 2 high.

Quick check
What is the period of y = sin(3x)?
  1. 2π/3, about 2.09
  2. 3
  3. π/3
  4. 3 × 2π
Show the answer

2π/3, about 2.09

The period is 2π/k = 2π/3. A bigger k means a shorter period: three cycles fit where one used to.

Quick check
Which wave has the tallest peaks: 4 sin(x), sin(4x) or sin(x + 4)?
  1. 4 sin(x)
  2. sin(4x)
  3. sin(x + 4)
  4. All three peak at 4
Show the answer

4 sin(x)

Only 4 sin(x) multiplies the height: its peaks are 4. sin(4x) repeats faster and sin(x + 4) is shifted, but both still peak at 1.

In your own words
In your own words: what do A, k and φ each change about a wave?

Write a sentence or two, then compare it with a good answer in the lesson.

In Python: a wave as a list of samples

Purpose. Turn a wave into numbers a program can store and use.

import math

samples = []
for i in range(8):
    x = i * 2 * math.pi / 8
    samples.append(round(math.sin(x), 3))
print(samples)
It runs right here in your browser, with nothing to install.

Each x is one eighth of a cycle further on, and each sample is the wave's height there. The list rises to 1, falls through 0 to −1, and comes back. Change the 8 to sample more or fewer points per cycle.

Code flow. Here is the flow's wave, y = 2 sin x, sampled by the same loop. Step through it and watch x move on an eighth of a cycle each time round, and samples grow by one box: the eight numbers the flow wrote, rounded to 3 places.

Too few samples lose the wave. Sample the same wave with fewer and fewer points per cycle:

import math

for n in [2, 4, 8]:
    samples = []
    for i in range(n):
        samples.append(round(math.sin(i * 2 * math.pi / n), 3))
    print(n, samples)

With 8 samples the shape is clear. With 4 it's a rough zig-zag. With 2 both samples land where the wave crosses 0, and the wave vanishes entirely. A rule of thumb: you need more than 2 samples per cycle of the fastest wave you want to keep. That's why audio is recorded tens of thousands of times a second.

Draw it. With plenty of samples, the joined points look like the smooth wave. Given one list, plt.plot puts each sample at its place in the list: sample 0, sample 1, and so on along the bottom. This cell takes 41 samples of y = 2 sin x, 40 even steps across one cycle, so sample 40 lands where the next cycle starts:

import math
import matplotlib.pyplot as plt

ys = []
for i in range(41):
    ys.append(2 * math.sin(i * 2 * math.pi / 40))
plt.plot(ys)
plt.xlabel("sample number i")
plt.ylabel("y = 2 sin x")
plt.show()
Quick check
Read the plot. At which sample number does the wave reach its lowest point, −2?
  1. 40
  2. 30
  3. 20
  4. 10
Show the answer

30

The lowest point comes three quarters of the way through a cycle, and three quarters of 40 steps is 30. Sample 10, a quarter of the way, is the peak, 2.

Worked example

Worked example: two waves added

Add a slow wave and a fast wave at a few points, the way sounds mix.

import math

total = []
for i in range(8):
    x = i * 2 * math.pi / 8
    slow = math.sin(x)
    fast = 0.5 * math.sin(3 * x)
    total.append(round(slow + fast, 3))
print(total)
In the lesson you can change this and run it.

The slow wave completes one cycle across the 8 points; the fast one, with k = 3, completes three, at half the height. Adding them sample by sample gives a new list: the slow rise and fall, with a ripple riding on it. Sound works exactly like this: two notes played together are two waves added, and a recording stores only the sum.

Quick check
What does this print?
import math

print(round(3 * math.sin(math.pi / 2), 3), round(math.sin(math.pi / 2 + math.pi / 2), 3))
  1. 3.0 1.0
  2. 3.0 0.0
  3. 1.0 0.0
  4. 3.0 -1.0
Show the answer

3.0 0.0

3 × sin(π/2) = 3 × 1 = 3.0, the peak. Shifting by another π/2 reaches π, where the wave is back at 0.

Quick check
Which line computes 8 samples across one cycle of sin x?
  1. i / 8 for i from 0 to 7
  2. i * 8 for i from 0 to 7
  3. i * 2 * math.pi / 8 for i from 0 to 7
  4. i * math.pi for i from 0 to 7
Show the answer

i * 2 * math.pi / 8 for i from 0 to 7

One cycle is 2π long, so 8 even steps are 2π/8 apart: i × 2π/8.

Drill

Sample a wave

Fix the loop so wave holds 12 samples of y = 2 sin(x) across one cycle, rounded to 3 places.

import math

wave = []
for i in range(12):
    x = i * 2 * math.pi / 8
    wave.append(round(math.sin(x), 3))
print(wave)
In the lesson, your code is checked as you write it.
In your own words
Why did sampling sin x at 2 points per cycle make the wave vanish?

Write a sentence or two, then compare it with a good answer in the lesson.

In AI and ML: sounds and positions

Purpose. Waves reach models in two ways.

Sound is a list of samples. A microphone measures air pressure thousands of times a second, and a speech model receives exactly that: a long list of wave heights, like the lists above. This lesson's sampling rule is why speech is usually recorded at 16,000 samples a second or more.

Positions as waves. A transformer reads all the words of a sentence at once, so it needs another way to know their order. The original transformer gave position p a list of wave values at several speeds:

code(p)=(sin⁡ps1,cos⁡ps1,sin⁡ps2,cos⁡ps2,…)\text{code}(p) = \left(\sin\tfrac{p}{s_1}, \cos\tfrac{p}{s_1}, \sin\tfrac{p}{s_2}, \cos\tfrac{p}{s_2}, \ldots\right)

In plain words: each position gets the heights of several waves, some fast and some slow. Fast waves tell neighbors apart; slow waves tell the start from the end.

Quick check
What does this print?
import math

p = 1
print(round(math.sin(p / 1), 3), round(math.sin(p / 10), 3))
  1. 0.1 0.841
  2. 0.841 0.841
  3. 1.0 0.1
  4. 0.841 0.1
Show the answer

0.841 0.1

At scale 1 the wave moves fast: sin(1) ≈ 0.841. At scale 10 it's slow: sin(0.1) ≈ 0.1. One position, two different wave values.

Why several speeds? One fast wave repeats every 2π ≈ 6.28 positions, so positions 0 and 6 get almost the same fast values. A slow wave changes too little to tell neighbors apart. Together they give every position its own pattern, the way a clock's second, minute and hour hands together give every moment its own reading.

Here are the codes for the first 8 positions: a fast wave (p ÷ 1) and a slow one (p ÷ 10), a sin and a cos of each.

import math

for p in range(8):
    code = []
    for s in [1, 10]:
        code.append(round(math.sin(p / s), 2))
        code.append(round(math.cos(p / s), 2))
    print(p, code)

Each line is one position's code: four numbers, each between −1 and 1, however far into the sentence the position is. The picture below stands each line on end, as a column. Read across a row to follow one wave from position 0 to 7.

Drill

Wave codes for positions

Set fast and slow to lists of sin(p / 1) and sin(p / 10) for positions 0 to 5, each rounded to 3 places. Then set climb to how far the slow wave rises from position 0 to position 5: slow[5] - slow[0], rounded to 3 places.

import math

fast = []
slow = []
for p in range(5):
    fast.append(round(math.sin(p * 1), 3))
    slow.append(round(math.sin(p * 10), 3))
climb = 0
print(fast, slow, climb)
In the lesson, your code is checked as you write it.

In the real tools. In PyTorch, the library most models are built with, the whole table comes from one line: a column of positions times a row of speeds gives every position at every speed, and torch.sin takes the sine of all of them at once. Here are your table's two sin rows, sin p and sin(p/10), with each position on a line of its own:

In your own words
Why does a transformer use waves of several speeds for positions, instead of just the position number?

Write a sentence or two, then compare it with a good answer in the lesson.

Goes deeper in

Practice what you just read

  • 5 code examples you can edit and run
  • a practice exercise that checks your code
  • a 5-question quiz
  • 9 review cards for spaced repetition

Try a quiz question

What are the amplitude and period of y = 3 sin(2x)?

  1. Amplitude 3, period π
  2. Amplitude 2, period 3
  3. Amplitude 3, period 2π
  4. Amplitude 6, period π

Answer it at the end of the lesson to see if you're right, and why.

Before this lesson

Keep learning

Where this lesson leads.

575 lessons in thirteen phases. 176 are free without an account, and a free account opens the rest. See every lesson.

Foundation · about 96 hours of study

Foundation math

From whole numbers to the math inside a neural network: each idea by hand, in Python, and in AI 118 lessons All free
  1. Numbers and place value FreeWhat a digit's place is worth, the number line, and Python as a calculator that never tires.36 min
  2. Multiplying and dividing: groups FreeEqual groups, quotient and remainder, and cutting training data into batches.35 min
  3. Order of operations, and names for numbers FreeBrackets, powers, then multiply and divide, then add and subtract; variables hold numbers.36 min
  4. Numbers below zero FreeThe number line through zero, sign rules, distance from zero, and an error's direction and size.37 min
  5. Fractions: equal parts, equal fractions, adding FreePieces of a whole, equal fractions, and re-cutting to add; accuracy as a fraction.45 min
  6. Multiplying and dividing fractions FreeA share of a share, dividing by flipping, and why chances in a row get small.35 min
  7. Decimals and rounding FreeFractions in place value, rounding, and why 0.1 + 0.2 isn't exactly 0.3.36 min
  8. Percentages FreePer hundred, percent change, percentage points, and the accuracy and error rates of models.35 min
  9. Ratios, rates and proportion FreeSharing in a ratio, rates per unit, scaling, and splitting data 80/10/10.35 min
  10. Powers and roots FreeRepeated multiplying, the laws of exponents, square roots, and squared error.43 min
  11. Very big and very small numbers FreePowers of ten, e-notation like 3e-4, orders of magnitude, and a model's size in memory.36 min
  12. Bits, bytes and floats: how numbers are stored FreeBase 2, what n bits can hold, a float as sign, power and digits, and a tensor's memory.35 min
  13. Review: find the slip (numbers) FreeBroken calculations to fix, mixed practice from the whole module, and its formula sheet.34 min
  14. Letters for numbers FreeVariables, terms and like terms, substitution, and a model's formula as a Python function.41 min
  15. Expanding and factoring FreeBrackets as areas, pulling out common factors, and a squared error expanded into a bowl.41 min
  16. Solving equations: keep the balance FreeThe same move on both sides, undoing steps in reverse, and running a model's formula backward.41 min
  17. Rearranging a formula FreeMaking any letter the subject, round-trip checks, and planning a training run.36 min
  18. Inequalities and true-or-false FreeRanges, flipping the sign, if and elif, and the thresholds, clips and ReLU inside a model.42 min
  19. Functions: a rule from input to output Freef(x) notation, tables and graphs, domain and range, loops over lists, and loss as a function.42 min
  20. Chaining and undoing functions Freeg(f(x)), inverses that undo, scaling and unscaling, and a network as a chain.42 min
  21. Straight lines: slope and intercept FreeRise over run, the line through two points, and a model's weight and bias as slope and intercept.42 min
  22. Two equations, two unknowns FreeSubstitution and elimination, where lines cross, and why models are fitted, not solved.42 min
  23. Quadratics: bowls and their lowest point FreeRoots by factoring and by formula, the discriminant, the vertex, and the best weight of a model.42 min
  24. Exponentials: growth and decay FreeThe same factor each step, the number e, and decaying learning rates.43 min
  25. Logarithms: the power question FreeTo what power?, the laws of logs, and adding log-probabilities instead of multiplying.45 min
  26. Sequences and Σ FreeArithmetic and geometric sequences, Σ as a loop that adds, and mean squared error.40 min
  27. Weighted and moving averages FreeWeights that add to 1, a window average, the exponential moving average, and smoothing a loss curve.35 min
  28. Sets, and reading the symbols FreeSets, union and intersection, ℝⁿ and f : ℝⁿ → ℝᵐ, and reading a paper's first equation.41 min
  29. Review: find the slip (algebra) FreeBroken algebra and code to fix, mixed practice from the whole module, and its formula sheet.34 min
  30. Transposing step by step: undo in reverse order FreeA formula's layers, the opposite of each, peeling from the outside in, conditions, and round-trip checks.35 min
  31. Formulas with fractions: clear them first FreeClearing denominators, cross-multiplying, a letter in the bottom, and reciprocal formulas like 1/R = 1/R₁ + 1/R₂.35 min
  32. When the letter appears twice: collect and factor FreeGathering the letter on one side, factoring it out, dividing by the bracket, and rearranging precision and F1.35 min
  33. Squares, roots and powers FreeSquare roots with ±, squaring to undo a root, cube and nth roots, and the √d inside attention.40 min
  34. When the letter is squared and plain: quadratics FreeRearranging into ax² + bx + c = 0, the formula, completing the square, and choosing the root that fits.35 min
  35. Letters in the power: exponentials and logarithms FreeTaking logs to bring a power down, exponentiating to free a letter, and the logit that undoes the sigmoid.40 min
  36. Rearranging inequalities: when the sign flips FreeThe four moves on an inequality, negatives and reciprocals that flip it, logs below 1, and sample sizes.35 min
  37. Inverse functions, lines and scaling FreeUndoing a function by making x the subject, domains that allow it, line forms, and unscaling predictions.40 min
  38. Sequences and sums: find n, a, d and r FreeEvery letter of the arithmetic and geometric formulas as the subject, and an average's window from its decay.40 min
  39. Combining formulas: substitute and eliminate FreePutting one formula inside another to remove a letter, and sizing a training run from a time or money budget.35 min
  40. Review: rearrange anything (algebra) FreeBroken rearrangements to fix, mixed practice from the whole algebra block, a strategy chart, and its formula sheet.35 min
  41. Coordinates and distance FreePoints as (x, y), Pythagoras, distance in any number of coordinates, and nearest neighbors.42 min
  42. Angles, degrees and radians FreeMeasuring a turn, converting units, arcs, and the angle between two arrows.36 min
  43. Sine and cosine FreeAcross and up: right triangles, the unit circle, and directions and rotations.42 min
  44. Waves: height, speed and shift FreeAmplitude, period and phase, sampling a wave, and position codes in transformers.37 min
  45. Review: find the slip (shapes and waves) FreeBroken geometry and code to fix, mixed practice from the module, and its formula sheet.29 min
  46. Angles, triangles and waves: rearranging with sin, cos and tan FreeSides and angles from SOH CAH TOA, principal values, arcs, polar coordinates, and a wave's time and period.45 min
  47. Vectors: lists of numbers that point FreeA list, a point and an arrow; components, dimension, NumPy arrays, and data as vectors.42 min
  48. Adding and scaling vectors FreeComponent-wise sums, scalar multiples, mixes, and the weight update w − ηg.43 min
  49. Length and distance: norms FreeL2, L1 and max norms, unit vectors, distance, and the norms inside losses and clipping.42 min
  50. The dot product FreeMultiply in pairs and add; a·a as squared length; a @ b; one neuron as a dot product.43 min
  51. Angle, cosine similarity and projection FreeDirection without length, perpendicular vectors, shadows, and search by meaning.45 min
  52. Perpendicular directions: bases and coordinates FreeOrthonormal directions, coordinates as dot products, rebuilding a vector, and taking a direction out.35 min
  53. Matrices: tables of numbers FreeRows, columns and entries; two-dimensional arrays and slices; a data set as a matrix.41 min
  54. Adding, scaling and transposing matrices FreeEntry-by-entry sums and scalings, the transpose, mean images, and tables turned around.36 min
  55. A matrix times a vector FreeEach row dotted with the vector, the size rule, A @ x, and a layer acting on one example.42 min
  56. Matrix multiplication FreeRow meets column, the size rule, order matters, and a batch through a layer: X W + b.45 min
  57. Identity, diagonal, triangular and symmetric FreeFour patterns and their jobs: rescaling, distance tables, and the causal mask.37 min
  58. Matrices as movements of space FreeStretches, flips, shears and rotations; columns as landing spots; chaining; a layer moving data.42 min
  59. Determinant and inverse FreeThe area factor ad − bc, singular matrices, undoing a matrix, and redundant features.42 min
  60. Solving systems by elimination FreeAugmented matrices, row moves, back-substitution, the three outcomes, and fitting weights.45 min
  61. Independence, span and rank FreeRedundant vectors, span, rank, outer products, and low-rank updates like LoRA.42 min
  62. Eigenvectors: directions that only stretch FreeAv = λv, checking candidates, np.linalg.eig, and the main direction of a data cloud.41 min
  63. SVD and PCA: the important directions first FreeA matrix as pieces by size, the best rank-k copy, compression, and PCA maps of data.45 min
  64. Tensors, shapes and broadcasting FreeAxes and shapes, reshape, axis=, broadcasting, and shapes through a small model.43 min
  65. One-hot vectors and lookup tables: embeddings FreeA one-hot row picks a table's row, a sentence looked up as vectors, and a score for every word.35 min
  66. Sliding dot products: convolution FreeA small pattern slid along a list and over a grid, output sizes, stride and padding, and shared weights.40 min
  67. Review: find the slip (vectors and matrices) FreeShape slips to fix, mixed practice from the module, and its formula sheet with shapes.36 min
  68. Vector equations: rearranging with arrows FreeSolving for a vector or a scale, recovering a gradient from two weight snapshots, and an angle from a dot product.40 min
  69. Matrix equations: multiply by the inverse, on the correct side FreeNo dividing by a matrix: inverses from the left or right, order, transposes, and the normal equations.40 min
  70. Rates of change FreeChange over change, secants, rates from readings with np.diff, and reading a loss curve.41 min
  71. Limits: what a value heads toward FreeApproaching from both sides, holes, corners and jumps, epsilon, and judging activations.37 min
  72. The derivative: slope at a point FreeShrinking secants, f′(x) and dy/dx, Δy ≈ f′Δx, and which way a weight should move.45 min
  73. Derivative rules FreeConstants, powers, multiples, sums, eˣ and ln x, checked numerically, and a squared error's slope.41 min
  74. The product rule and the chain rule FreeProducts and chains of functions, forward then backward, and one neuron's dL/dw.45 min
  75. Lowest points and curvature Freef′ = 0, the second-derivative test, tables, local and global minima, and convex losses.43 min
  76. Standing in for a curve near a point FreeTangent lines, a curvature term, growing error, and why learning rates must be small.37 min
  77. Partial derivatives FreeOne input at a time, ∂, nudging in Python, and the two slopes of a line's loss.41 min
  78. The gradient FreeAll the slopes as a vector, steepest climb and descent, a numerical gradient, and the loss gradient.42 min
  79. Gradient descent FreeStepping against the gradient, the learning rate, a training loop, and fitting a line.45 min
  80. Faster descent: momentum and Adam FreeMomentum and Adam by hand, the bias correction, and three optimizers down one narrow valley.40 min
  81. The chain rule with many paths: backpropagation by hand FreeGraphs, multiply along paths and add across them, gradient checks, and a 2-2-1 network by hand.45 min
  82. Automatic differentiation: how PyTorch finds every slope FreeA tape of steps walked backward, and requires_grad, backward(), .grad, zero_grad() and no_grad().40 min
  83. Jacobians and Hessians FreeSlopes of vector functions, a layer's Jacobian, second partials, and bowls, domes and saddles.42 min
  84. Integrals: adding up thin slices FreeStrip and trapezoid sums, summing undoes differencing, probability as area, and AUC.43 min
  85. Review: find the slip (calculus) FreeCalculus slips to fix, mixed practice from the module, and its formula sheet.36 min
  86. Rearranging in calculus: slopes, steps and constants FreeSetting a slope to zero, Newton's step, the chain rule turned around, constants from a point, and learning rates.45 min
  87. Counting: choices, orders and groups FreeThe counting principle, factorials, ordered picks and groups, and why attention's cost grows with the square of the length.43 min
  88. Probability: from counts to chances FreeEqually likely outcomes, the complement, distributions, seeded simulation, and a next-word model.43 min
  89. Combining events: and, or, independent FreeThe or rule, independence, probability trees, drawing with and without replacement, and chains of steps.42 min
  90. Conditional probability and Bayes' rule FreeConditionals from tables, why P(A | B) isn't P(B | A), Bayes' rule, and precision and recall.45 min
  91. Random variables and distributions FreeUncertain numbers, the probability of a range, sampling by slices, and sampling versus argmax.42 min
  92. Expected value, variance and standard deviation FreeProbability-weighted averages, spread in the data's own units, and a training loss as an expected value.42 min
  93. Bernoulli, binomial, uniform and normal FreeYes-or-no trials, counts of yeses, the 68-95-99.7 rule, z-scores, and an accuracy's error bar.45 min
  94. Two variables at once: covariance and correlation FreeJoint tables and marginals, covariance, correlation, the covariance matrix, and features that carry the same information.45 min
  95. Review: find the slip (probability) FreeProbability slips to fix, mixed practice from the module, and its formula sheet.36 min
  96. Probability formulas: solve for the missing chance FreeConditionals, Bayes' rule for the prior, the or rule, expected values, variance, and z-scores turned around.45 min
  97. Describing data: center, spread and scale FreeMean, median and mode, variance and standard deviation, percentiles, outliers, and standardizing features.42 min
  98. Samples, the law of large numbers and the central limit theorem FreeSamples and populations, the standard error σ/√n, the bell of averages, and error bars on test scores.43 min
  99. Is the difference real? Comparing two models FreePaired differences, a bootstrap interval, what a p-value is and isn't, and why many tries make lucky wins.40 min
  100. Maximum likelihood FreeLikelihoods as products, log-likelihoods as sums, grid search, underflow, and the negative log-likelihood loss.45 min
  101. Linear regression: the best line FreeResiduals, least squares by formula, three roads to one line, the matrix form, and train versus test error.45 min
  102. Logistic regression: a line that answers yes or no FreeThe sigmoid, the log loss, its slope p − y, a training loop, and choosing a threshold.45 min
  103. Scoring a classifier: confusion table, precision, recall, F1 FreeThe four cells, accuracy, precision, recall and F1, and what moving a threshold does to each.35 min
  104. Softmax: scores into probabilities FreeExponentiate and share out, shifting and temperature, the safe softmax, and a classifier's last step.43 min
  105. Surprise, entropy and cross-entropy FreeSurprise in bits, entropy, cross-entropy and KL divergence, perplexity, and a language model's loss.45 min
  106. KL divergence: how far one distribution is from another FreeKL by hand, cross-entropy as entropy plus KL, why it isn't symmetric, and a soft-target loss.35 min
  107. Review: find the slip (statistics and information) FreeStatistics slips to fix, mixed practice from the module, and its formula sheet.38 min
  108. Statistics and information: n, z, logits and bits FreeSample sizes from error bars, inputs from probabilities, logit gaps from softmax, and chances from bits and losses.40 min
  109. One neuron, start to finish FreeWeighted sum, activation, loss, slopes and update, by hand and in code, and what one straight line can and can't split.45 min
  110. A layer and a small network as matrices FreeTwo layers on a batch, the shape of every matrix, adding a hidden unit, and XOR solved.45 min
  111. Training a small network FreeThe backward pass in matrix form, gradient checks, a full training loop, and reading loss curves.45 min
  112. The same network in PyTorch Freenn.Sequential with the NumPy weights loaded, the same loss and slopes, then SGD and mini-batches.45 min
  113. Attention as matrix math FreeScaled dot-product scores, the causal mask, a row-by-row softmax, and blending value vectors.45 min
  114. Residual connections and layer norm FreeAdding the input back and why its slope keeps learning alive, layer norm by hand, and add and norm.40 min
  115. A transformer block, start to finish FreeA sentence through embeddings, attention, add and norm, a small network and next-word scores, every shape named.45 min
  116. Planning a model: rearranging the formulas of a working network FreeLayer widths from parameter counts, sizes from memory and compute, the attention scale, and a training plan.40 min
  117. Final test: transpose any formula FreeSlips to fix from every topic, mixed practice of every kind of rearranging, the strategy chart and the full sheet.45 min
  118. Capstone: a sign reader from the formulas FreeBuild, train and test a trail-sign classifier with NumPy alone, and report its accuracy with an error bar.45 min
Part of a free course 118 lessons

The Foundation Math For Ai & ML: from arithmetic to attention

Learn the math behind AI from arithmetic up: numbers, algebra, vectors and matrices, calculus, probability and statistics, each shown as numbers moving and checked in Python, ending with a network you build and train from the formulas.

Never trained an AI? Teach a small one to write, right in your browser. It takes about 20 seconds and needs no account. Teach it to write