102013
Algebra · Concept hubDeep story

Vectors and Matrices

1844 CE19th-century Germany and Ireland (Grassmann and Hamilton)

Concept

Mathematics that tables relationships and transforms directions, spaces, and data together. Chinese equation tables, Japanese and European determinant traditions, and nineteenth-century vector spaces and matrix algebra grew from different problems before becoming important computational languages in graphics, quantum theory, statistics, search, and machine learning.

Understand it in one breath

A vector packages components into one object, and a matrix compactly represents a linear transformation. Cayley's 1858 memoir helped develop earlier matrix calculations into a general algebra. Matrices are central to graphics, neural networks, quantum mechanics, and statistics, but a single matrix does not represent every nonlinear transform, and translations require an extension such as homogeneous coordinates.

At a glance

Matrix

Meaning

Effect on a unit square

[[2,0],[0,2]]

Scale by 2

Area ×4

[[cos θ, -sin θ],[sin θ, cos θ]]

Rotation (θ)

Area preserved

[[-1,0],[0,1]]

Reflection across the y-axis

Area preserved; orientation reversed

[[1,1],[0,1]]

Shear

Area preserved; square becomes a parallelogram

[[a,b],[c,d]]

General transformation

Area ×|ad−bc|

A matrix is a compact representation of a spatial transformation. The absolute determinant |det| is the area scale factor; a negative determinant reverses orientation.

Key formula

(Ax)i=jAijxj(A\mathbf{x})_i = \sum_{j} A_{ij}\, x_j

Key moments

263 CE

Commenting on the Nine Chapters — eliminating a whole table

Liu Hui’s 263 commentary explained why counting-rod tabular procedures for simultaneous equations worked. It was not an abstract modern matrix, but it preserves a long history of arranging and reducing relationships together.

1683 CE

Seki’s determinant-like calculations

Seki Takakazu presented determinant-like rules within Japan’s wasan tradition, independently of Leibniz’s European path. This was not yet all of modern linear algebra.

1843 CE

Hamilton’s quaternions — a precursor to vectors

Hamilton used a four-dimensional quaternion system to algebraize directions and rotations. It was not today’s vector notation, but became one of several shoulders for vector analysis.

1844 CE

Grassmann’s exterior algebra

German schoolteacher Hermann Grassmann published a general theory of extension quantities in arbitrary dimensions. Unfamiliar notation slowed reception, but the work became a foundation for vector spaces and exterior algebra.

1858 CE

Cayley’s matrix algebra

Working as a lawyer in London, Cayley systematized matrix addition, multiplication, and inverses after Sylvester’s 1850 naming and earlier determinant and transformation work.

1925 CE

Matrix mechanics — multiplication order becomes physics

Born recognized matrices in Heisenberg’s transition arrays, and work with Jordan and then all three developed noncommutative matrix mechanics. It was a collaborative development.

2017 CE

The Transformer — making contextual comparison matrix-shaped

Parallel attention through query, key, and value matrices helped scale translation and generative models. Matrix multiplication alone does not explain language understanding or all of AI.

Modern applications

Matrix operations express many linear and homogeneous-coordinate transformations in graphics, neural-network layers, PageRank, and PCA. They are a major computational tool, while nonlinear models and complete systems require much more than matrices alone.

Beyond MathVoyage

Loading…