Forward Selection for Time Series-Based Qubit Generation via Parameterized Quantum Gates

Quantum data processing requires classical data to be encoded into quantum states. Current noisy intermediate-scale quantum devices have a limited number of qubits that are stable only briefly. Encoding classical data into qubits is the initial step in Quantum Machine Learning (QML), and effective encoding is crucial for quantum processing. This algorithms for data processing are still emerging, and compact data representations are essential for their success. This research proposes a novel data encoding technique using uniformly controlled rotation gates, achieving high storage density by encoding real-valued time series data as qubit rotations. The model uses a binary representation for computations on time series data, reducing the number of quantum measurements needed. The research explores quantum forward propagation in simulations to improve prediction accuracy for time series signals using parameterized quantum circuits, handling trends, noise, and sinusoidal components. The efficiency of the encoding process depends on data volume and chosen encoding, with potential infinite loading time in the worst case. This study presents a Forward Selection Time Series Data Pro-cessing and Feature Extraction Model for Qubits generation with Parameterized Quantum Gates (FSDPFEM-PQG), demonstrating superior performance in quantum representations compared to existing models.