SHENZHEN, China, Jan. 5, 2026 /PRNewswire/ -- MicroCloud Hologram Inc. (NASDAQ:HOLO), ("HOLO" or the "Company"), a technology service provider, released learnable quantum spectral filter technology for hybrid graph neural networks. This achievement proposes a brand-new quantum-classical hybrid graph neural network foundational architecture. By mapping the graph Laplacian operator to a trainable quantum circuit, it enables graph signal processing to gain exponential compression capability and a new computational perspective, representing a key step for quantum graph machine learning toward practicalization.

HOLO's this technology proposes a quantum spectral filter that fuses graph convolution and pooling operations into a complete quantum computing process. The input signal is loaded into the quantum state using amplitude encoding or probability encoding. The quantum circuit performs spectral transformation based on the graph structure. After passing through learnable rotation gates and controlled gates, the measurement results of the output state naturally form an n-dimensional probability distribution vector, where n = log(N). This property enables the quantum circuit to directly map high-dimensional graph signals to low-dimensional space, achieving a unified function of convolution + pooling.

HOLO points out that the quantum measurement process is essentially a structured nonlinear mapping, capable of overcoming the complex structural search problems in classical GNN pooling operations. In quantum circuits, nonlinear behaviors that are difficult to simulate in classical networks are automatically realized through quantum state collapse, making the pooling results both compressive and separable while preserving key spectral features of the graph structure.

This means that a graph of size N, after processing through the quantum convolution layer, can immediately obtain log(N)-dimensional compressed features, with computational costs remaining controllable even for large graphs. For a network with one million nodes, classical spectral convolution is almost impossible to run in terms of memory and time, whereas this quantum circuit requires only about 20 qubits.

The mathematical foundation of this technology stems from the spectral structure of the graph Laplacian operator. The Laplacian operator L = D - A has a natural coupling relationship with the graph structure, and its eigenvalues reflect important properties such as graph connectivity, clustering structure, and smoothness. Traditional graph neural networks utilize the eigenvalues of L to filter signals, but spectral computation must rely on complex numerical linear algebra.

HOLO proves that through the QFT-structured quantum circuit, the feature space of graphs can be approximated. This conclusion relies on two key discoveries: first, an effective mapping can be constructed between the graph's adjacency matrix and quantum gates—by building controlled rotation gates corresponding to graph edges, the coupling structure of the circuit simulates local adjacency relationships on the graph; second, the hierarchical rotation logic in QFT naturally contains a multi-scale filtering structure, consistent with the decoupling capability of high-frequency and low-frequency components in the graph spectrum. When the depth of the quantum circuit is designed to be polynomial level, it is only necessary to trainably adjust the rotation angles and phases to approximate the eigenbasis of the Laplacian matrix.

To reduce the number of qubits, HOLO adopts a spectral approximation method based on logarithmic encoding, that is, representing the original N-dimensional feature space using n = log(N) qubits. The Hilbert space dimension constructed by this method is 2^n, theoretically capable of one-to-one mapping with the N-dimensional space.

In engineering implementation, the training of the quantum circuit is completed through classical-quantum hybrid optimization. The classical optimizer computes the gradients of the loss function with respect to circuit parameters and calculates the differentiability of the quantum circuit through the parameter shift rule. The quantum circuit extracts spectral features from high-dimensional input encoded signals and outputs low-dimensional features that can be further processed by classical networks. The entire system forms an end-to-end trainable hybrid GNN.

Large-scale graph learning has always been a difficult problem in the industrial field. Domains such as social media, traffic flow networks, and internet connectivity graphs each have tens of millions or even hundreds of millions of nodes. Classical GNNs typically require large amounts of video memory, long-duration matrix multiplications, complex sparse matrix management, and massive convolution filter parameters.

In contrast, quantum spectral filters provide a disruptive solution. As the number of nodes grows exponentially, the required qubits grow only logarithmically, making it a natural choice for future quantum-classical GNNs. Particularly in the current stage where quantum hardware is about to enter the medium-scale phase, this method with low qubit demand and high structural utilization offers excellent implementation possibilities.

HOLO believes that rather than waiting for the full maturation of quantum hardware, it is more important to build quantum frontier algorithm infrastructure in advance. This quantum spectral filter has established a complete research route, deeply integrating graph structures with quantum learnable models, laying an algorithmic foundation for future hardware development.

With the official release of HOLO's learnable quantum spectral filter for hybrid graph neural networks, the fusion of quantum computing and graph neural networks has taken a key step forward. HOLO not only demonstrates the enormous potential of quantum circuits in complex structure learning but also opens up a practical and scalable technical path for future quantum machine learning.

The successful implementation of this technology is driving graph neural networks toward a true quantum era. In the future, as quantum hardware gradually matures, such learnable quantum filters will become core components in numerous practical applications, constituting a brand-new cornerstone for the integrated development of graph computing, artificial intelligence, and physical computing.

About MicroCloud Hologram Inc.

MicroCloud Hologram Inc. (NASDAQ:HOLO) is committed to the research and development and application of holographic technology. Its holographic technology services include holographic light detection and ranging (LiDAR) solutions based on holographic technology, holographic LiDAR point cloud algorithm architecture design, technical holographic imaging solutions, holographic LiDAR sensor chip design, and holographic vehicle intelligent vision technology, providing services to customers offering holographic advanced driving assistance systems (ADAS). MicroCloud Hologram Inc. provides holographic technology services to global customers. MicroCloud Hologram Inc. also provides holographic digital twin technology services and owns proprietary holographic digital twin technology resource libraries. Its holographic digital twin technology resource library utilizes a combination of holographic digital twin software, digital content, space data-driven data science, holographic digital cloud algorithms, and holographic 3D capture technology to capture shapes and objects in 3D holographic form. MicroCloud Hologram Inc. focuses on developments such as quantum computing and quantum holography, with cash reserves exceeding 3 billion RMB, and plans to invest more than 400 million in USD from the cash reserves to engage in blockchain development, quantum computing technology development, quantum holography technology development, and derivatives and technology development in frontier technology fields such as artificial intelligence AR. MicroCloud Hologram Inc.'s goal is to become a global leading quantum holography and quantum computing technology company.

Safe Harbor Statement

This press release contains forward-looking statements as defined by the Private Securities Litigation Reform Act of 1995. Forward-looking statements include statements concerning plans, objectives, goals, strategies, future events or performance, and underlying assumptions and other statements that are other than statements of historical facts. When the Company uses words such as "may," "will," "intend," "should," "believe," "expect," "anticipate," "project," "estimate," or similar expressions that do not relate solely to historical matters, it is making forward-looking statements. Forward-looking statements are not guarantees of future performance and involve risks and uncertainties that may cause the actual results to differ materially from the Company's expectations discussed in the forward-looking statements. These statements are subject to uncertainties and risks including, but not limited to, the following: the Company's goals and strategies; the Company's future business development; product and service demand and acceptance; changes in technology; economic conditions; reputation and brand; the impact of competition and pricing; government regulations; fluctuations in general economic; financial condition and results of operations; the expected growth of the holographic industry and business conditions in China and the international markets the Company plans to serve and assumptions underlying or related to any of the foregoing and other risks contained in reports filed by the Company with the Securities and Exchange Commission ("SEC"), including the Company's most recently filed Annual Report on Form 10-K and current report on Form 6-K and its subsequent filings. For these reasons, among others, investors are cautioned not to place undue reliance upon any forward-looking statements in this press release. Additional factors are discussed in the Company's filings with the SEC, which are available for review at www.sec.gov. The Company undertakes no obligation to publicly revise these forward-looking statements to reflect events or circumstances that arise after the date hereof.

View original content:https://www.prnewswire.com/news-releases/microcloud-hologram-inc-releases-learnable-quantum-spectral-filter-technology-for-hybrid-graph-neural-networks-302652751.html

SOURCE MicroCloud Hologram Inc.

MicroCloud Hologram Inc. Releases Learnable Quantum Spectral Filter Technology for Hybrid Graph Neural Networks