Forensics of Transpiled Quantum Circuits

TL;DR

This work tackles the problem of backend transparency in cloud-based quantum computing by proposing a forensics approach that derives the hardware coupling map from transpiled circuits. The method hinges on detecting SWAP-based routing and constructing a swap-history signature to infer the effective coupling graph, then combining results across many circuits to reconstruct the backend topology. Experiments on IBM backends with linear, T-shaped, H-shaped, and loop couplings show high accuracy: the coupling map can be extracted with complete accuracy for almost all cases, and up to $97.33\%$ of programs traced to the correct backend; full backend maps can be recovered with modest numbers of circuits, even in mixed-backend pools. This work demonstrates a practical step toward trust and accountability in quantum cloud services and provides a foundation for forensics in future quantum tooling.

Abstract

Many third-party cloud providers set up quantum hardware as a service that includes a wide range of qubit technologies and architectures to maximize performance at minimal cost. However, there is little visibility to where the execution of the circuit is taking place. This situation is similar to the classical cloud. The difference in the quantum scenario is that the success of the user program is highly reliant on the backend used. Besides, the third-party provider may be untrustworthy and execute the quantum circuits on less efficient and more error-prone hardware to maximize profit. Thus, gaining visibility on the backend from various aspects will be valuable. Effective forensics can have many applications including establishing trust in quantum cloud services. We introduce the problem of forensics in the domain of quantum computing. We trace the coupling map of the hardware where the transpilation of the circuit took place from the transpiled program. We perform experiments on various coupling topologies (linear, T-shaped, H-shaped, and loop) on IBM backends. We can derive the coupling map from the transpiled circuits with complete accuracy for almost every transpiled circuit we considered. We could correctly trace 97.33% of the programs to the correct backend.

Figures (8)

• Figure 1: Derivation of the physical coupling map from a transpiled circuit.
• Figure 2: (1) Coupling map for IBM Kyiv. The qubits in red are an example of the linear topology, the green ones - an H-shaped topology, blue signifies the loop topology and pink is an example of the T-shaped topology. (2) Coupling map of IBM Cambridge obtained by running the analysis on the four subgraphs showed using different colors. (3) T-shaped coupling map which is identical for IBM Burlington and IBM Vigo.
• Figure 3: Percentage of fake_cambridge backend extracted with respect to the number of transpiled circuits of different sizes.
• Figure 4: Percentage of fake_singapore backend extracted with respect to the number of transpiled circuits of different sizes.
• Figure 5: Percentage of backend predicted using pools of differently sized transpiled circuits for (a) fake_singapore (b) fake_cambridge.