WiMi Releases Hybrid Quantum-Classical Neural Network (H-QNN) Technology for Efficient MNIST Binary Image Classification
Rhea-AI Summary
WiMi (NASDAQ:WIMI) released a Hybrid Quantum-Classical Neural Network (H-QNN) for efficient MNIST binary image classification on Feb 6, 2026. The H-QNN uses a parameterized quantum circuit for feature encoding, quantum-state feature extraction, and a classical MLP classifier with hybrid optimization. WiMi reports ~30% lower simulation compute time versus comparable classical networks and observed improved feature expressivity when scaling qubits from 4 to 8.
The framework targets extensibility to handwriting recognition, medical imaging, and video-frame feature extraction, and WiMi plans device-level verification and broader quantum-algorithm integration.
Positive
- Simulation time reduced by ~30% in simulation versus classical networks
- Scalability signal observed when qubits increased from 4 to 8
Negative
- None.
Market Reaction
Following this news, WIMI has gained 12.64%, reflecting a significant positive market reaction. Our momentum scanner has triggered 12 alerts so far, indicating notable trading interest and price volatility. The stock is currently trading at $2.05. This price movement has added approximately $2M to the company's valuation.
Data tracked by StockTitan Argus (15 min delayed). Upgrade to Silver for real-time data.
Key Figures
Market Reality Check
Peers on Argus
WIMI was down 11.22% while peers were mixed: ABLV +3.89%, ACCS -5.11%, FLNT -5.19%, MCTR +0.65%, SWAG -2.22%, indicating stock-specific dynamics rather than a sector-wide move.
Previous AI Reports
| Date | Event | Sentiment | Move | Catalyst |
|---|---|---|---|---|
| Jan 05 | AI model announcement | Positive | +10.9% | Launch of MC-QCNN quantum CNN for multi-channel supervised learning. |
| Dec 22 | Hybrid QNN launch | Positive | -1.4% | Next-gen hybrid quantum neural network for image multi-classification. |
| Dec 22 | Hybrid QNN launch | Positive | -1.4% | Same H-QNN multi-classification announcement with negative price reaction. |
| Oct 23 | SHQCNN research | Positive | +5.9% | Shallow hybrid quantum-classical CNN model to improve image classification. |
| Oct 20 | SQ-QNN development | Positive | +2.1% | Single-qudit quantum neural network for multi-class and multi-task design. |
AI/quantum announcements have usually led to positive moves, though some events showed negative reactions despite upbeat technical news.
Over recent months, WiMi has repeatedly highlighted quantum-enhanced AI. On Oct 20, 2025, it introduced single-qudit SQ-QNN technology, followed by SHQCNN research on Oct 23, 2025. A next‑gen hybrid H‑QNN for multi‑classification arrived on Dec 22, 2025, then MC‑QCNN on Jan 5, 2026. These AI-tagged releases typically produced moves around ±3.22%, with most events seeing positive 24‑hour reactions to technical progress updates.
Historical Comparison
Past AI-tagged quantum AI announcements moved WIMI shares by an average of 3.22%, framing this H-QNN MNIST update within an ongoing quantum-AI strategy.
AI-tagged history shows progression from single-qudit and shallow hybrid CNNs to multi-class H-QNNs and MC-QCNN, with this MNIST-focused H-QNN extending the quantum-classical framework to binary image classification.
Market Pulse Summary
The stock is surging +12.6% following this news. A strong positive reaction aligns with WiMi’s history of AI-tagged quantum releases, which previously moved shares around 3.22% on average. The H-QNN MNIST work reinforces a pattern of frequent quantum-AI updates, but past events also show occasional negative reactions. With shares trading well below the 200-day MA 3.5 and near the 52-week low 1.81 beforehand, investors would have weighed whether enthusiasm for technical progress could persist once initial excitement faded.
Key Terms
hybrid quantum-classical neural network technical
convolutional neural networks (cnn) technical
multi-layer perceptrons (mlp) technical
hilbert space technical
cnot technical
AI-generated analysis. Not financial advice.
Beijing, Feb. 06, 2026 (GLOBE NEWSWIRE) -- WiMi Releases Hybrid Quantum-Classical Neural Network (H-QNN) Technology for Efficient MNIST Binary Image Classification
BEIJING, Feb.06, 2026––WiMi Hologram Cloud Inc. (NASDAQ: WiMi) ("WiMi" or the "Company"), a leading global Hologram Augmented Reality ("AR") Technology provider, today announced the release of a Hybrid Quantum-Classical Neural Network (Hybrid Quantum-Classical Neural Network, H-QNN) technology for efficient MNIST binary image classification. This breakthrough achievement marks a new progress in quantum machine learning moving from theoretical exploration toward practicalization, and also embodies the enterprise's core competitiveness in the field of quantum intelligent algorithm research. This technology takes an efficient hybrid structure, scalable quantum feature mapping mechanism, and quantum state optimization strategy as its core, successfully achieving excellent classification performance on the MNIST handwritten digit dataset, proving the practical feasibility and computational advantages of quantum neural networks in high-dimensional image recognition tasks.
Under the traditional framework of deep learning, convolutional neural networks (CNN) have long played the core role in image feature extraction, while multi-layer perceptrons (MLP) undertake the final classification tasks. However, these models are still constrained by the bottlenecks of classical computing architectures, especially in the complex feature mapping and nonlinear discrimination of high-dimensional data, where models are prone to issues such as overfitting, gradient vanishing, and high computational complexity. The emergence of quantum computing provides new solutions to this problem. Quantum neural networks (QNN) can represent complex feature distributions in an exponentially large Hilbert space by leveraging quantum superposition and entanglement characteristics, thereby theoretically achieving feature expression capabilities far surpassing those of classical neural networks.
The hybrid quantum-classical neural network (H-QNN) technology proposed by WiMi was born precisely under this technological trend. H-QNN introduces a trainable quantum feature encoding module at the front end of the classical network, mapping raw image data into a high-dimensional quantum feature space, then performing nonlinear feature transformations using quantum circuits, and finally conducting subsequent classification decisions through the classical network. This structure fully combines the exponential expressive power of quantum computing in feature mapping with the mature mechanisms of classical deep learning in large-scale parameter optimization, achieving synergistic enhancement between quantum and classical computing. Unlike pure quantum networks or classical deep models, H-QNN not only avoids the limitations of high noise and limited qubit numbers in quantum hardware but also retains the potential acceleration advantages of quantum algorithms in feature extraction.
From the perspective of architectural design, the H-QNN technology consists of three main parts: the data preprocessing module, the quantum encoding and feature extraction module, and the classical neural classifier. First, the data preprocessing module performs binarization and normalization operations on the 28×28 pixel images of MNIST, and reduces the image dimensionality to a quantumizable data format through compression and block strategies. At this stage, WiMi uses a screening method based on statistical feature distribution to ensure that the data input into the quantum system has high feature representativeness, thereby reducing the generation of invalid quantum states.
After entering the quantum encoding stage, H-QNN adopts a Parameterized Quantum Circuit (PQC) as the core computing unit. The PQC is composed of several layers of quantum gates, which structurally include rotation gates (R_y, R_z) and entanglement gates (CNOT, CZ) and other operations, used to construct nonlinear quantum feature space mappings. Image pixels or locally extracted feature vectors are mapped to quantum states, embedding numerical information into quantum amplitudes or phases through quantum rotation encoding. This process achieves the quantum expression of high-dimensional nonlinear features, making each sample possess a unique global representation in the quantum state space.
In the core stage of feature extraction, H-QNN simulates complex high-dimensional decision boundaries through quantum state evolution. The superposition and entanglement properties of quantum states enable the model to simultaneously capture multiple feature correlation relationships in a single evolution. Through the iterative action of multi-layer quantum circuits, the quantum states of the input data are mapped to new feature distributions, and the output results, after measurement, yield a set of interpretable feature vectors. Unlike traditional convolutional or fully connected layers, this quantum transformation layer can theoretically achieve exponential feature space expansion without significantly increasing the number of parameters.
The measurement results serve as intermediate feature vectors input into the classical neural classifier part. This part adopts a lightweight multi-layer perceptron structure, composed of several fully connected layers and nonlinear activation functions. Through classical backpropagation algorithms, the model can simultaneously update the quantum circuit parameters and classical weights, thereby achieving hybrid optimization. To maintain the training stability of the quantum and classical parts, WiMi introduces a hybrid optimization strategy based on gradient estimation. This strategy precisely calculates the gradients of trainable parameters in the quantum circuit through the Parameter Shift Rule, thereby ensuring the differentiability and convergence of the overall network during the training process.
- QNN demonstrates significant performance advantages in experiments. In the binary classification task of the MNIST dataset, WiMi uses the distinction between handwritten digits "0" and "1" as the experimental basis. Experimental results indicate that H-QNN's classification accuracy under the same number of training epochs and sample scale is significantly higher than that of classical MLP models of equivalent scale. More importantly, the introduction of the quantum feature space significantly enhances the model's sensitivity and discrimination ability to high-dimensional features; even under smaller sample sets, the model still maintains excellent generalization performance. This characteristic indicates that the feature mapping mechanism of the quantum part effectively reduces overfitting phenomena and enhances robustness to noise and anomalous data.
In addition to the improvement in performance metrics, the computational efficiency of H-QNN has also been fully verified. Due to the parallelism characteristics of quantum circuits, the computation time of the model in the simulation environment is reduced by approximately
The H-QNN technology is not merely a classification model for MNIST but a general quantum-enhanced neural network framework. The design concept of this framework can be extended to more computer vision tasks, including handwriting recognition, medical image analysis, and even video frame feature extraction fields. By adjusting the quantum encoding method and circuit depth, the model can adapt to datasets of different dimensions and noise levels, providing novel efficient learning solutions for enterprise-level AI applications.
WiMi plans to further verify the operability and noise resistance performance of H-QNN technology on actual quantum devices. At the same time, it will also explore the integration with other quantum algorithms (such as quantum support vector machines and quantum convolutional networks) to build a more general quantum intelligence framework. In addition, quantum feature compression and distributed quantum learning for large-scale visual datasets are also important directions for the next stage of research. It is believed that, with the continuous advancement of quantum hardware, hybrid quantum-classical neural networks will become an important pillar in the evolution of artificial intelligence computing architectures.
WiMi's hybrid quantum-classical neural network (H-QNN) for efficient MNIST binary image classification is not only a technological innovation but also an important milestone in quantum intelligence moving toward real-world applications. It proves that efficient, stable, and synergistic integration can be achieved between quantum computing and deep learning, showcasing the infinite potential of future quantum machine learning in high-dimensional data analysis, image understanding, and pattern recognition fields. WiMi will continue to be committed to promoting research in the foundational theory and application technologies of quantum artificial intelligence, laying a solid foundation for the future of intelligent computing.
About WiMi Hologram Cloud
WiMi Hologram Cloud Inc. (NASDAQ: WiMi) focuses on holographic cloud services, primarily concentrating on professional fields such as in-vehicle AR holographic HUD, 3D holographic pulse LiDAR, head-mounted light field holographic devices, holographic semiconductors, holographic cloud software, holographic car navigation, metaverse holographic AR/VR devices, and metaverse holographic cloud software. It covers multiple aspects of holographic AR technologies, including in-vehicle holographic AR technology, 3D holographic pulse LiDAR technology, holographic vision semiconductor technology, holographic software development, holographic AR virtual advertising technology, holographic AR virtual entertainment technology, holographic ARSDK payment, interactive holographic virtual communication, metaverse holographic AR technology, and metaverse virtual cloud services. WiMi is a comprehensive holographic cloud technology solution provider. For more information, please visit http://ir.wimiar.com.
Translation Disclaimer
The original version of this announcement is the officially authorized and only legally binding version. If there are any inconsistencies or differences in meaning between the Chinese translation and the original version, the original version shall prevail. WiMi Hologram Cloud Inc. and related institutions and individuals make no guarantees regarding the translated version and assume no responsibility for any direct or indirect losses caused by translation inaccuracies.
Investor Inquiries, please contact:
WIMI Hologram Cloud Inc.
Email: pr@wimiar.com
ICR, LLC
Robin Yang
Tel: +1 (646) 975-9495
Email: wimi@icrinc.com
FAQ
What is WiMi's H-QNN and how does it apply to MNIST (WIMI)?
How much faster is WiMi's H-QNN compared with classical models (WIMI)?
What accuracy improvements did WiMi report for MNIST binary classification (WIMI)?
Can WiMi's H-QNN be extended beyond MNIST (WIMI)?
What are WiMi's next steps for H-QNN development (WIMI)?
