WiMi Releases Next-Generation Hybrid Quantum Neural Network Structure Technology, Breaking Through the Bottleneck of Image Multi-Classification

Core Viewpoint - WiMi Hologram Cloud Inc. has launched a hybrid quantum neural network structure (H-QNN) that integrates classical convolutional neural networks (CNN) with quantum neural networks (QNN) to enhance image multi-classification capabilities, achieving superior classification accuracy and stability compared to existing algorithms [1][13]. Group 1: Technology Overview - The H-QNN design combines classical abstraction with quantum discrimination, consisting of three main modules: feature dimensionality reduction and encoding, quantum state transformation, and hybrid decision and transfer learning [2]. - The feature dimensionality reduction module utilizes CNN to extract low-dimensional features, employing an improved angle encoding method to efficiently map features to quantum states [3]. - The quantum state transformation module performs high-dimensional feature mapping and nonlinear discrimination, using parameterized rotation gates and controlled entanglement gates to maintain gradient balance during training [4]. - The hybrid decision and transfer learning module integrates quantum computing results with classical decision layers, enhancing generalization performance through a transfer learning mechanism [5]. Group 2: Innovation Points - The architectural design achieves deep integration of CNN and QNN, moving beyond traditional models by adopting a three-stage structure for feature extraction, quantum mapping, and decision-making [8]. - The encoding strategy employs a joint dimensionality reduction scheme of angle encoding and PCA, addressing quantum encoding dimension limitations and maximizing information fidelity [9][10]. - The training strategy introduces a transfer learning mechanism and parameter sharing to mitigate risks of gradient vanishing and overfitting, along with an early stopping strategy based on quantum fidelity metrics [11]. Group 3: System Implementation - The system utilizes a heterogeneous computing architecture, running classical components on CPU/GPU platforms while executing quantum parts in FPGA-based quantum simulation modules, significantly improving training speed [12]. - This hybrid computing architecture demonstrates performance advantages over pure CPU or GPU simulations, marking a significant advancement in practical applications of quantum artificial intelligence [13]. Group 4: Company Background - WiMi Hologram Cloud Inc. specializes in holographic cloud services, focusing on areas such as in-vehicle AR holographic HUD, 3D holographic pulse LiDAR, and metaverse holographic technologies, positioning itself as a comprehensive solution provider in the holographic AR technology sector [14].