Speaker
Description
We present a novel approach to embed node-attributed graphs onto neutral-atom quantum computers, which extends PASQAL's quantum evolution kernel (QEK) by introducing a node-level feature encoding based on three-level quantum states. Despite recent advances, the incorporation of multiple continuous node attributes into analog quantum graph kernels remains a challenging and mostly unexplored problem.
Our primary objective is to assess the architectural feasibility of multi-attribute continuous encoding on analog hardware, providing a baseline for future large-scale implementations. We achieve this by encoding node-level features via local Raman laser pulses, exploiting the two hyperfine ground states alongside the Rydberg state to form a three-level (qutrit) system per atom.
To construct our kernel, we adopt the Wasserstein distance between probability distributions—computed from correlation and occupation observables—in place of the more commonly used Jensen–Shannon metric, which proved less effective according to our experimental results on the digit recognition task.
| Topic | Quantum kernel methods and their applications |
|---|---|
| Additional Topic | Encoding and processing of data in quantum systems |