New methods for encoding large data into quantum states

Researchers developed scalable ways to put complex data into quantum systems using fewer gates.

New methods allow for deterministic and scalable amplitude-encoding of large data. These methods use low two-qubit gate costs. They offer tunable trade-offs between circuit depth and encoding fidelity.

The approach uses two complementary algorithms to simplify multiplexer circuits. One method maps the elimination of redundant gate controls to a Boolean logic minimization problem. The second method uses a single ancilla qubit to map multi-controlled rotation gates to one- and two-qubit operations via the Walsh-Hadamard spectrum of the gate angles.

Amplitude-encoding is a method for putting data into quantum states. A circuit depth refers to the number of operations in a sequence. Fidelity is a measure of how accurate the state is. An ancilla qubit is an extra qubit used in a circuit.

The methods were tested on IQM hardware with a square lattice connectivity. Hybridizing these methods can improve circuit depth by up to orders of magnitude compared to the default method in the Qiskit package. This reduction in gate depth helps prevent noise from ruining calculations before they start.

These methods can encode arbitrary complex data or queryable functions. They use tensor-networks to go beyond what can be held in classical memory. This work may shorten the time horizon to robust quantum utility.

Key facts

  • The methods require no amplitude amplification or mid-circuit measurement.
  • Hybridizing the algorithms provides up to orders of magnitude improvement in circuit depth.
  • The approach was demonstrated on IQM hardware with excellent fidelity.

All sections

Edition