Classification Using a Two-Qubit Quantum Chip

Hrant Gharibyan, Hovnatan Karapetyan, Tigran Sedrakyan +4 more

We perform image classification on the Honda Scenes Dataset on Quantinuum's H-2 and IBM's Heron chips utilizing up to 72 qubits and thousands of two-qubit gates. For data loading, we extend the hierarchical learning to the task of approximate amplitude encoding and block amplitude encoding for commercially relevant images up to 2 million pixels. Hierarchical learning enables the training of variational circuits with shallow enough resources to fit within the classification pipeline. For comparison, we also study how classifier performance is affected by using piecewise angle encoding. At the end of the VQC, we employ a fully-connected layer between measured qubits and the output classes. Some deployed models are able to achieve above 90\% accuracy even on test images. In comparing with classical models, we find we are able to achieve close to state of the art accuracy with relatively few parameters. These results constitute the largest quantum experiment for image classification to date.

Shuang Cong, Jingru Qiu

Emmanuel Zambrini Cruzeiro, Christine De Mol, Serge Massar +1 more

Kai Huang

Heuristic classification of the single-molecule carbene spin-photon interface reported by Roggors et al. (2026). Three labels organise the published numbers: stored pattern, selection rule, and projection channel. Comparison measures are arithmetic from published numbers only. No new measurement is reported. No information field is calculated and no coupling value is fitted. Entanglement, a two-qubit gate, and on-chip integration were not shown in the experiment and are not claimed here.

Hsiao-Chung Lin, Hui Zhang, Hong Nian Cai +2 more

On the silicon photonic chip, we apply Mach Zehnder Interferometer (MZI) and optical network to manipulate the phase difference and amplitude distribution of the input light. The two or three-dimensional input data is encoded in the light amplitude distribution and the genetic algorithm is applied to update the angle parameters of the MZIs to improve the classification result.

Joanna Wiśniewska, Marek Sawerwain