Many scientific and engineering problems reduce to a system of linear equations:

A quantum linear-systems algorithm aims to prepare a quantum state proportional to the solution vector (\mathbf{x}). A variational quantum linear algorithm approaches this using a parameterised quantum circuit and a classical optimiser.

The name is sometimes written VQLA or VQLS, depending on the paper or implementation. The central hybrid idea is the same.

The target

Encode the right-hand side as a quantum state (|b\rangle). We want to prepare a state (|x\rangle) satisfying

Rather than constructing (|x\rangle) directly, choose a parameterised circuit

The parameters (\theta) are adjusted until the circuit output behaves like the solution.

The hybrid loop

1. Choose circuit parameters (\theta).
2. Prepare (|x(\theta)\rangle).
3. Estimate a cost function measuring how poorly it satisfies the linear system.
4. Send the cost to a classical optimiser.
5. Update (\theta) and repeat.

A natural cost is related to the residual

Practical formulations rewrite this into expectation values that quantum circuits can estimate.

Why is it interesting?

Linear systems appear throughout simulation, optimisation, differential equations, and machine learning. A useful quantum method could therefore have wide reach.

But there are major qualifications: loading (|b\rangle), representing (A), choosing an expressive ansatz, avoiding optimisation failure, controlling measurement cost, and deciding which useful property of the solution should be extracted.

What implementation teaches

Coding the method forces several questions:

• How is (A) decomposed into measurable operators?
• How is (|b\rangle) prepared?
• Which ansatz and cost function are used?
• How is correctness tested on a small classical problem?
• What is measured after optimisation?

These details reveal what the algorithm really assumes.

Recording version

A variational quantum linear algorithm guesses a quantum state for the solution, measures how badly it fails the equation, and uses a classical optimiser to improve the guess.

Draw a loop with four objects: parameterised circuit, candidate state, cost function, and classical optimiser.