Roberta Zambrini

Quantum kernel methods leverage a kernel function computed by embedding input information into the Hilbert space of a quantum system. However, large Hilbert spaces can hinder generalization capability, and the scalability of quantum kernels becomes an issue. To overcome these challenges, various strategies under the concept of inductive bias have been proposed. Bandwidth optimization is a promising approach that can be implemented using quantum simulation platforms. We propose trapped-ion simulation platforms as a means to compute quantum kernels and demonstrate their effectiveness for binary classification tasks. We compare the performance of the proposed method with an optimized classical kernel and evaluate the robustness of the quantum kernel against noise. The results show that ion trap platforms are well-suited for quantum kernel computation and can achieve high accuracy with only a few qubits.

Non-homogeneous chain environments (e.g. segmented ion traps) are investigated through an exact diagonalization approach. Different spectral densities, including band-gaps, can be engineered to separately assess memory effects. Environment non-Markovianity is quantified with recently introduced measures of information flow-back and non-divisibility of the system dynamical map. By sweeping the bath spectrum via tuning of the system frequency we show strongest memory effect at band-gap edges and provide an interpretation based on energy flow between system and environment. A system weakly coupled to a stiff chain ensures a Markovian dynamics, while the size of the environment as well as the local density of modes are not substantial factors. We show an opposite effect when increasing the temperature inside or outside the spectral band-gap. Further, non-Markovianity arises for larger (negative and positive) powers of algebraic spectral densities, being the Ohmic case not always the most Markovian one.