qaoa

Documentation

Overview

Package qaoa implements the Quantum Approximate Optimization Algorithm.

QAOA solves combinatorial optimization problems by alternating cost and mixer layers parameterized by gamma and beta angles.

Index

• func MaxCutHamiltonian(edges [][2]int, numQubits int) (pauli.PauliSum, error)
• type Config
• type Result
  • func Run(ctx context.Context, cfg Config) (*Result, error)

Functions

func MaxCutHamiltonian

func MaxCutHamiltonian(edges [][2]int, numQubits int) (pauli.PauliSum, error)

MaxCutHamiltonian encodes a graph as a cost Hamiltonian for the MaxCut problem. C = 0.5 * sum_{(i,j) in edges} (I - Z_i*Z_j) Minimizing -C maximizes the cut. The returned Hamiltonian is negated so that minimization yields the maximum cut.

Types

type Config

type Config struct {
	// CostHamiltonian is the objective to minimize.
	CostHamiltonian pauli.PauliSum
	// MixerHamiltonian is the mixer operator. Nil uses the default X-mixer.
	MixerHamiltonian *pauli.PauliSum
	// Layers is the number of QAOA layers (p). Default: 1.
	Layers int
	// Optimizer is the classical optimization method.
	Optimizer optim.Optimizer
	// InitialParams are the starting [gamma_1..gamma_p, beta_1..beta_p]. Nil means heuristic.
	InitialParams []float64
	// Shots for sampling the best bitstring. Default: 1024.
	Shots int
}

Config specifies the QAOA problem and solver.

type Result

type Result struct {
	OptimalValue  float64
	OptimalParams []float64
	BestBitstring string
	BestCost      float64
	NumIters      int
	Converged     bool
}

Result holds QAOA output.

func Run

func Run(ctx context.Context, cfg Config) (*Result, error)

Run executes QAOA.