New method for simulating complex quantum systems

Researchers developed a technique called patch routing to make quantum simulations more efficient.

A new method called patch routing helps create efficient block encodings for frustrated lattice Hamiltonians. These Hamiltonians are built from repeated local interaction patches.

The process first block-encodes a small local patch. A SWAP-style routing network then applies this encoding across the lattice. Block-encoding is a core primitive in many quantum algorithms, including quantum simulation.

The expensive local block-encoding cost is paid only once per query. This cost is independent of lattice size. The routing network grows linearly with the lattice size and dominates at large sizes.

This method works for the Heisenberg, Fermi-Hubbard, and honeycomb spin-liquid models. For a 120×120120\times 120 square-lattice J1J_1-J2J_2 Heisenberg model, it reduces the leading TT cost per qubitized-QPE run by 15−17×15-17\times.

The method also provides a 3×3\times reduction for the 120×120120\times 120 Fermi-Hubbard model at half-filling. For the honeycomb spin-liquid model, the leading cost is reduced by 5×5\times for 120×120120\times 120 unit cells.

This framework provides a systematic way forward for larger and more accurate fault-tolerant simulations of real-world quantum systems.

Key facts

  • Patch routing reduces the leading TT cost for a 120×120120\times 120 square-lattice J1J_1-J2J_2 Heisenberg model by 15−17×15-17\times.
  • The local block-encoding cost is independent of lattice size.
  • The method uses a SWAP-style routing network to apply local patch encoding across a lattice.

All sections

Edition