analiza aiqbench
This paper studies how connectivity in IQP (Instantaneous Quantum Polynomial-time) circuits affects their use as variational ansätze for Hamiltonian optimization. Its main claim is a trade-off: denser connectivity increases entanglement and expressibility but tends to shrink gradient variance, while sparser designs keep gradients stable. For readers tracking near-term variational quantum algorithms, the practical takeaway is that more expressive is not automatically better, and that a circular topology offered the best balance on the authors' benchmarks. The evidence is limited to the paper's own benchmarks, so the generality of these findings still needs independent testing.
Discovery of connectivity-trainability trade-off of IQP circuits for Hamiltonian optimization
arXiv:2606.24264v3 Announce Type: replace-cross Abstract: Instantaneous Quantum Polynomial-time (IQP) circuits are promising candidates for near-term quantum computing due to the conjectured classical hardness of their sampling task and their shallow, highly parallelizable structure. Despite extensive studies of their complexity-theoretic properties, the role of circuit connectivity in their trainability as variational ansatze remains poorly understood. In this work, we present an investigation of IQP circuits for Hamiltonian optimization that traces a mechanistic chain linking circuit connectivity, entanglement generation, and gradient scaling: increasing connectivity drives the circuit output toward Haar-like statistics and volume-law entanglement, and this increase in entanglement is consistently associated with an exponential decay of the gradient variance and degraded optimization performance, while sparser architectures retain stable gradients at the cost of global expressibility and entanglement capacity. This connectivity-entanglement-gradient mechanism produces a nontrivial connectivity-trainability trade-off, demonstrating that effective IQP-based Hamiltonian optimization requires a balance between global expressibility and trainability rather than maximal expressivity alone. In particular, on our benchmarks, a circuit generating no entanglement already matches the optimization performance of more highly connected circuits, and a circular topology retains entanglement capacity while preserving trainability, making it a broadly effective choice among the architectures considered.
Abstract:Instantaneous Quantum Polynomial-time (IQP) circuits are promising candidates for near-term quantum computing due to the conjectured classical hardness of their sampling task and their shallow, highly parallelizable structure. Despite extensive studies of their complexity-theoretic properties, the role of circuit connectivity in their trainability as variational ansatze remains poorly understood. In this work, we present an investigation of IQP circuits for Hamiltonian optimization that traces a mechanistic chain linking circuit connectivity, entanglement generation, and gradient scaling: increasing connectivity drives the circuit output toward Haar-like statistics and volume-law entanglement, and this increase in entanglement is consistently associated with an exponential decay of the gradient variance and degraded optimization performance, while sparser architectures retain stable gradients at the cost of global expressibility and entanglement capacity. This connectivity-entanglement-gradient mechanism produces a nontrivial connectivity-trainability trade-off, demonstrating that effective IQP-based Hamiltonian optimization requires a balance between global expressibility and trainability rather than maximal expressivity alone. In particular, on our benchmarks, a circuit generating no entanglement already matches the optimization performance of more highly connected circuits, and a circular topology retains entanglement capacity while preserving trainability, making it a broadly effective choice among the architectures considered.
Social and Information Networks (cs.SI); Numerical Analysis (math.NA); Data Analysis, Statistics and Probability (physics.data-an); Quantum Physics (quant-ph)
arXiv:2606.24264 [cs.SI]
(or arXiv:2606.24264v3 [cs.SI] for this version)
https://doi.org/10.1088/1402-4896/aeb6c6 DOI(s) linking to related resources
Text integral sub licența deschisă a publicației · 220 cuvinte