Matrices, graph theory and Markov-chain methods are especially natural within IB Mathematics AI HL because they turn relationships, networks and state changes into mathematical models.
Matrices
Matrices can represent transformations, systems and transitions. Students should understand the meaning of rows, columns and multiplication rather than treating matrix operations as isolated calculator commands.
Graph theory
Vertices and edges provide a language for networks. Degree, paths, cycles, connectivity and weighted networks can model transport, communication and other real systems.
Markov chains
A transition matrix describes how probability moves between states. Repeated multiplication reveals short-run and long-run behaviour, but students should always ask whether the Markov assumption is plausible for the context.
2029-course caution
The revised AI course begins first teaching in August 2027 and first assessment in May 2029. IB has published targeted reductions to HL content. Students must check the applicable course generation rather than carrying every current HL topic forward automatically.
For stable matrix and network Mathematics beyond the IB syllabus, use BTT’s advanced Mathematics libraries.
World Mathematics route: return to the World Mathematics Atlas for the wider map across examinations, curricula, competitions, mathematical objects and university routes.
IB Mathematics AI HL: matrices, graph theory and Markov chains as modelling languages
Matrices organise transformations and systems
A matrix can encode data, linear relationships or transformations. Operations should be interpreted from the model, not treated as arbitrary table arithmetic.
Matrix multiplication is order-sensitive
AB and BA need not be equal. Dimensions determine whether a product exists, and the order often corresponds to composition of processes.
Inverse matrices solve reversible linear systems
Where an inverse exists, it can recover unknowns or reverse a transformation. Determinant or other course conditions indicate when inversion is impossible.
Graphs model networks
Vertices and edges can represent locations, people, routes or dependencies. Degree, paths, cycles and connectivity answer questions that ordinary coordinate graphs do not.
Weighted graphs add cost or distance
When edges carry weights, route problems can represent distance, time or cost. The numerical optimum still needs contextual interpretation.
Algorithms need traceable steps
For shortest paths, spanning structures or other syllabus algorithms, record the state changes systematically. A correct final network without enough process can hide an invalid choice.
Markov chains model state transitions
A transition matrix records probabilities of moving among states under the model. Rows or columns must follow a consistent convention.
Long-run behaviour depends on the transition structure
Repeated multiplication can reveal stabilising distributions where the model permits. Interpret the limiting vector in context rather than reporting calculator output alone.
Assumptions matter
Markov models assume the next transition depends on the current state according to the specified probabilities. Real systems may violate stationarity or memory assumptions.
Technology supports scale
GDC or software can multiply matrices and iterate transitions quickly, but students must construct the matrix correctly and explain what entries mean.
Common errors
Reversing matrix order, mixing row/column conventions, confusing graph-theory graphs with function graphs and interpreting a steady-state vector as certainty are common.
Transfer
Use transport, migration, market-share or network contexts so the student learns to construct the model from prose rather than only execute a supplied matrix.