Index Ai Academy
Foundation · Vectors and matrices

The dot product

Multiply in pairs and add; a·a as squared length; a @ b; one neuron as a dot product.

About 43 minutes · Free lesson

What it covers6

Why it matters

The dot product turns two lists of numbers into one number: multiply them pair by pair and add. It's a weighted total, like the bill for a shopping basket, amounts times prices. It's also the single most important operation in machine learning. One neuron computes exactly one dot product, a layer computes many, and a transformer spends most of its time doing them. This lesson builds it by hand, three ways in Python, and as a neuron in NumPy and PyTorch. It also meets two facts that the rest of the module leans on: a vector dotted with itself gives its squared length, and two vectors whose dot product is 0 meet at a right angle.

What you'll be able to do

  • Compute a dot product: multiply matching components and add
  • Read it as a weighted total, like a shopping bill, and as a shadow whose sign says whether two arrows lean together; know that a·a is the squared length
  • Compute it three ways in Python, a loop, sum(a * b) and a @ b, and check the lengths first
  • See one neuron's calculation, z = w·x + b, as a dot product plus a bias, and get the same z from PyTorch's nn.Linear
Quick check
Have a guess before you start. You buy 3 apples at $1.50, 2 loaves at $4 and 5 lemons at $0.80. What's the bill?
  1. $16.50
  2. $10
  3. $6.30
  4. $63
Show the answer

$16.50

3 × 1.50 + 2 × 4 + 5 × 0.80 = 4.50 + 8 + 4 = 16.50. That multiply-and-add is a dot product of amounts and prices.

The math: multiply in pairs, then add

Purpose. One number that says how much two lists agree, weighted.

The dot product of two vectors of the same length multiplies matching components and adds the results. For a = (2, 3, 1) and b = (4, 1, 5):

a⋅b=2×4+3×1+1×5=16\mathbf{a} \cdot \mathbf{b} = 2 \times 4 + 3 \times 1 + 1 \times 5 = 16

In plain words: pair them up, multiply each pair, add the products.

The shopping bill is the everyday case: amounts · prices. The answer is a single number, a scalar, not a vector. The two vectors must be the same length, or some numbers have no partner.

Quick check
From an earlier lesson: what is Σ i² for i = 1 to 3?
  1. 6
  2. 9
  3. 36
  4. 14
Show the answer

14

1 + 4 + 9 = 14. A dot product is a Σ too: the sum over i of aᵢ × bᵢ.

The picture. Watch (2, 3, 1) · (4, 1, 5) two ways at once: as boxes of numbers, and as a drawing. In the drawing, a's numbers sit in the circles on the left and b's numbers ride on the lines. Each pair multiplies along its line, then the products pour into the circle on the right, marked Σ, which keeps the running total. That circle is one unit of a neural network, and the AI part comes back to it.

Try it now

Then set vector a to 1, 2 and vector b to 2, -1, and press Use these numbers. The two products cancel, and the dot product is 0.

Symbols and formula. The dot product, and two facts worth knowing:

a⋅b=∑i=1naibia⋅a=∥a∥2\mathbf{a} \cdot \mathbf{b} = \sum_{i=1}^{n} a_i b_i \qquad \mathbf{a} \cdot \mathbf{a} = \lVert \mathbf{a} \rVert^2

In plain words: a vector dotted with itself is its length squared: (3, 4) · (3, 4) = 9 + 16 = 25 = 5².

Order doesn't matter. a · b = b · a, because each product aᵢbᵢ is the same both ways round.

Zero means perpendicular. (1, 2) · (2, −1) = 2 − 2 = 0. When the dot product of two nonzero vectors is 0, they point at right angles: neither has any part along the other. Lesson 51 turns this into a measure of angle.

Reading the sign: shadows. A positive dot product means the two vectors lean the same way; a negative one means they lean apart. In two dimensions you can see it. Take b = (1, 0), an arrow one unit long along the x-axis, and shine a light straight down onto b's line. The arrow a casts a shadow on that line, starting at the origin (the picture draws it just under the line, so b stays in view). Because b is one unit long, a · b is exactly the shadow's length: positive when the shadow falls forward along b, negative when it falls behind.

Quick check
What is (1, 0, 2) · (3, 5, −1)?
  1. −2
  2. 8
  3. 1
  4. 3
Show the answer

1

1 × 3 + 0 × 5 + 2 × (−1) = 3 + 0 − 2 = 1.

Quick check
Which pair of vectors is perpendicular?
  1. (1, 1) and (1, 2)
  2. (2, 1) and (2, 1)
  3. (3, 0) and (2, 0)
  4. (2, 1) and (−1, 2)
Show the answer

(2, 1) and (−1, 2)

2 × (−1) + 1 × 2 = 0: a dot product of 0 means right angles.

In your own words
In your own words: why must two vectors be the same length to have a dot product?

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

In Python: three ways, one answer

Purpose. See the dot product as a loop, then use NumPy's shorthand.

import numpy as np

a = np.array([2, 3, 1])
b = np.array([4, 1, 5])
total = 0
for i in range(len(a)):
    total = total + a[i] * b[i]
print(total, np.sum(a * b), a @ b)
It runs right here in your browser, with nothing to install.

The loop is the definition written out. a * b multiplies matching components (lesson 48), and np.sum adds them: the same two steps in one line. The @ operator does both at once and is what real code uses. All three give 16.

Code flow. Step through the loop one line at a time. It runs on plain lists here; the loop works the same on arrays. Watch i pick each pair in turn, and total grow by that pair's product: the running sum from the picture's circle.

A named mistake: * for @. a * b alone is a vector of products, not their total. It runs without an error, so check what you get: a dot product is a single number.

import numpy as np

a = np.array([2, 3, 1])
b = np.array([4, 1, 5])
print((a * b).tolist(), a @ b)

Check the lengths first. @ on vectors of different lengths stops with an error. A good function says why, before NumPy does, by comparing len(a) with len(b).

Why three ways? The loop is the definition, step by step, and it's the version to write when you need to see what's happening. np.sum(a * b) shows the two stages, multiply then add, as two NumPy operations. And a @ b is the one real code uses: it's shortest and fastest, and lesson 56 will use the same @ to multiply whole matrices. All three must agree, which makes the first two a handy test of the third.

Worked example

Worked example: a shopping bill, three ways

Compute the bill for amounts (3, 2, 5) at prices (1.50, 4.00, 0.80) by loop, by sum, and with @.

import numpy as np

amounts = np.array([3, 2, 5])
prices = np.array([1.5, 4.0, 0.8])
bill = 0
for i in range(len(amounts)):
    bill = bill + amounts[i] * prices[i]
print(round(bill, 2), round(np.sum(amounts * prices), 2), round(amounts @ prices, 2))
In the lesson you can change this and run it.

All three give 16.5, the bill from the opening question. Each pair is an item: its amount times its price. The rounding tidies float noise in the money. In ML the "amounts" are a model's inputs and the "prices" its weights: a dot product says how much each input contributes, weighted by how much the model cares about it.

Quick check
What does this print?
import numpy as np

a = np.array([1, 2])
b = np.array([2, -1])
print(a @ b, a @ a)
  1. 0 5
  2. 0 3
  3. 5 0
  4. 4 5
Show the answer

0 5

1 × 2 + 2 × (−1) = 0, so they're perpendicular. a · a = 1 + 4 = 5, the squared length.

The shadows, in code. The three arrows from the math part's picture, each dotted with b = (1, 0):

import numpy as np

b = np.array([1, 0])
print(np.array([4, 2]) @ b, np.array([0, 3]) @ b, np.array([-3, 2]) @ b)

Each answer is a's first number: dotting with b = (1, 0) keeps the first number (times 1) and drops the second (times 0). That's the shadow's length with its sign: forward, none, behind. Change [4, 2] to [1, 5] and predict the first answer before you run it.

Quick check
Which line gives the dot product of a and b as one number?
  1. a * b
  2. (a * b).tolist()
  3. a + b
  4. a @ b
Show the answer

a @ b

@ multiplies matching components and adds them up.

Drill

Three ways to the same number

Fix dot(a, b) so it checks the lengths match (returning None if not) and then computes the dot product with a loop. Then set with_at to the same product with @.

import numpy as np

def dot(a, b):
    total = 0
    for i in range(len(a)):
        total = total + a[i] + b[i]
    return total

a = np.array([2, -1, 3])
b = np.array([1, 4, 2])
with_at = a * b
print(dot(a, b), with_at, dot(a, np.array([1, 2])))
In the lesson, your code is checked as you write it.
In your own words
Why is it worth writing the dot product as a loop once, even though @ exists?

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

In AI and ML: one neuron is one dot product

Purpose. The basic unit of a neural network computes a dot product plus a number.

A neuron takes an input vector x, multiplies it by its weight vector w, and adds a bias b:

z=w⋅x+bz = \mathbf{w} \cdot \mathbf{x} + b

In plain words: each input times its weight, all added up, plus the bias. That's lesson 14's ŷ = w·x + b, with many inputs instead of one.

import numpy as np

w = np.array([0.5, -1.0, 2.0])
x = np.array([4.0, 1.0, 3.0])
b = 0.5
print((w * x).tolist(), w @ x + b)

The products show each input's contribution: 2, −1 and 6, plus the bias 0.5, making z = 7.5. It's the drawing from the math part with a bias added at the end: the inputs in the circles, the weights on the lines, and z in the circle they feed.

Quick check
A neuron has weights (0.2, 3, −0.5) and gets the input (10, 1, 2). Which input contributes the most to z?
  1. The third: its weight is negative
  2. The second: 3 × 1 = 3
  3. They all contribute equally
  4. The first: it's the biggest input
Show the answer

The second: 3 × 1 = 3

The contributions are 0.2 × 10 = 2, 3 × 1 = 3 and −0.5 × 2 = −1. The second is biggest, because its weight is large.

Reading weights. A weight says how much the neuron cares about its input: a big positive weight makes that input push z up strongly, a negative one pushes it down, and a weight near 0 means the input hardly matters. Training adjusts these weights so the dot product comes out high for some inputs and low for others. A layer is many neurons side by side, which lesson 55 writes as one matrix times one vector.

The bias shifts the whole total. b doesn't multiply any input; it's added at the end, so it moves z up or down by the same amount for every input. Without it, an input of all zeros would always give z = 0. With it, the neuron can be "on" or "off" by default, before any input pushes it either way. Training learns b along with the weights: it's one more number that the update in lesson 48 nudges at every step.

Try both ideas on the neuron below. It has just two inputs, to keep it small: slide each input, each weight and the bias, and watch the sum change.

In PyTorch. The same neuron in the real tool, two ways. First, this lesson's line, w @ x + b, on tensors. Second, nn.Linear(3, 1): PyTorch's layer with 3 inputs and 1 output, which is one neuron with a weight for each input and one bias. A new layer starts with random weights, so the cell loads ours. copy_(w) copies w into the weights and fill_(b) sets the bias; a name ending in _ changes a tensor in place. with torch.no_grad(): tells PyTorch not to record the lines under it for training, which is how you set weights by hand. Without it, PyTorch refuses to change a layer's weights in place.

import torch
import torch.nn as nn

w = torch.tensor([0.5, -1.0, 2.0])
x = torch.tensor([4.0, 1.0, 3.0])
b = 0.5
print(w @ x + b)

layer = nn.Linear(3, 1)
with torch.no_grad():
    layer.weight.copy_(w)
    layer.bias.fill_(b)
print(layer(x).item())

Both give 7.5, the z the NumPy cell printed: first as a tensor, then as a plain number, which .item() takes out of a one-number tensor. In a real model, nobody loads the weights by hand. Training finds them, nudging every weight and the bias a little at each step.

Quick check
What does this print?
import numpy as np

w = np.array([1.5, -2.0, 0.5])
x = np.array([2.0, 1.0, 4.0])
print((w * x).tolist(), w @ x - 1)
  1. [3.0, -2.0, 2.0] 3.0
  2. [3.0, -2.0, 2.0] 2.0
  3. 3.0 [3.0, -2.0, 2.0]
  4. [3.5, -1.0, 4.5] 2.0
Show the answer

[3.0, -2.0, 2.0] 2.0

The contributions are 3, −2 and 2; their sum is 3, and the bias −1 makes z = 2.

Drill

One neuron's output

Write neuron(w, x, b) returning w · x + b. Then set z for weights (1.5, −2, 0.5), input (2, 1, 4) and bias −1, and biggest to the position of the input that contributes most.

import numpy as np

def neuron(w, x, b):
    return w * x + b

w = np.array([1.5, -2.0, 0.5])
x = np.array([2.0, 1.0, 4.0])
z = neuron(w, x, -1)
contrib = w * x
biggest = 0
print(z, contrib.tolist(), biggest)
In the lesson, your code is checked as you write it.
In your own words
What does a negative weight do to its input's effect on a neuron?

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

Goes deeper in

Practice what you just read

  • 6 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 is (2, 5) · (3, −1)?

  1. 1
  2. 11
  3. −1
  4. 6

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