Circuit cutting with classical side information
arXiv:2503.22384 · doi:10.1103/38mx-36k6
Abstract
Circuit cutting is a technique for simulating large quantum circuits by partitioning them into smaller subcircuits, which can be executed on smaller quantum devices. The results from these subcircuits are then combined in classical post-processing to accurately reconstruct the expectation value of the original circuit. Circuit cutting introduces a sampling overhead that grows exponentially with the number of gates and qubit wires that are cut. Many recently developed quasiprobabilistic circuit cutting techniques leverage classical side information, obtained from intermediate measurements within the subcircuits, to enhance the post-processing step. In this work, we provide a formalization of general circuit cutting techniques utilizing side information through quantum instruments. With this framework, we analyze the advantage that classical side information provides in reducing the sampling overhead of circuit cutting. Surprisingly, we find that in certain scenarios, side information does not yield any reduction in sampling overhead, whereas in others it is essential for circuit cutting to be feasible at all. Furthermore, we present a lower bound for the optimal sampling overhead with side information that can be evaluated efficiently via semidefinite programming and improves on all previously known lower bounds.
24 pages, 2 figures
References in corpus (20)
- Error mitigation for short-depth quantum circuits
- Practical Quantum Error Mitigation for Near-Future Applications
- Everything You Always Wanted to Know About LOCC (But Were Afraid to Ask)
- Application of a resource theory for magic states to fault-tolerant quantum computing
- Trading classical and quantum computational resources
- Simulating Large Quantum Circuits on a Small Quantum Computer
- Estimating outcome probabilities of quantum circuits using quasiprobabilities
- Doubling the size of quantum simulators by entanglement forging
- Robustness of Magic and Symmetries of the Stabiliser Polytope
- Circuit knitting with classical communication
- A framework for bounding nonlocality of state discrimination
- Constructing a virtual two-qubit gate by sampling single-qubit operations
- Combining quantum processors with real-time classical communication
- Overhead for simulating a non-local channel with local channels by quasiprobability sampling
- Randomly distilling W-class states into general configurations of two-party entanglement
- Optimal quantum circuit cuts with application to clustered Hamiltonian simulation
- Optimal Entangling Capacity of Dynamical Processes
- Cutting circuits with multiple two-qubit unitaries
- Simulating all multipartite non-signalling channels via quasiprobabilistic mixtures of local channels in generalised probabilistic theories
- Power of quantum measurement in simulating unphysical operations