Insider Brief
WiMi is exploring a multi-dimensional pooling optimization approach that combines variational quantum algorithms, the Quantum Haar Transform, and quantum partial measurement techniques.
The proposed framework is designed to preserve local feature information while reducing data dimensionality in high-dimensional datasets.
According to the company, the approach could support future quantum machine learning applications involving images, audio, point clouds, and hyperspectral data.
Photo from Unsplash by Dynamic Wang.
