MicroAlgo Inc. Develops Quantum Architecture Search (QAS) Technology to Enhance VQA Robustness and Trainability, Optimizing the Potential of Quantum Computing Devices

SHENZHEN, China, May 8, 2026 /PRNewswire/ -- MicroAlgo Inc. (the "Company" or "MicroAlgo") (NASDAQ: MLGO), today announced the development of an innovative technology—Quantum Architecture Search (QAS), aimed at automatically optimizing the architecture of quantum circuits to enhance the robustness and trainability of VQA, maximizing the potential of quantum computing devices.

In the traditional VQA framework, the design of quantum circuit architectures is typically performed manually or based on certain predefined standard architectures. However, the noise and errors in quantum computers are extremely severe in medium-scale devices, making circuit design a critical factor affecting VQA performance. More complex circuit architectures may enhance expressive power but simultaneously introduce more noise and errors, leading to difficulties in the training process or even complete failure.

To balance the expressive power of circuit architectures and the impact of noise, MicroAlgo has proposed a Quantum Architecture Search (QAS) method. QAS optimizes VQA performance by automatically searching for quantum circuit architectures, mitigating the impact of noise on training, and finding a near-optimal circuit structure. This method not only helps improve the robustness of quantum algorithms in noisy environments but also significantly enhances their performance in practical tasks.

The core idea of MicroAlgo QAS is to systematically search the architecture space of quantum circuits to find the circuit structure most suitable for a specific task. Unlike traditional design, QAS adopts an intelligent optimization approach, automatically exploring the space of circuit architectures to maximize the trainability and robustness of VQA.

The design of quantum circuit architectures is not merely a matter of arranging quantum gates; it involves multiple levels of optimization, such as the selection of quantum gates, the connectivity of qubits, and the interaction patterns between qubits. QAS first defines a circuit architecture space that encompasses all possible quantum circuit configurations, including the types, order, and connection patterns of quantum gates.

To effectively search the circuit architecture space, QAS introduces advanced optimization methods such as reinforcement learning and genetic algorithms. First, QAS uses a reinforcement learning model to evaluate the performance of VQA under different architectures by simulating the training process. Through this approach, QAS can select the optimal solution from millions of possible circuit architectures.

Additionally, noise in quantum computing is one of the key factors limiting VQA performance. During the architecture search process, QAS specifically incorporates a noise modeling mechanism, which predicts the performance of different circuit architectures under noisy conditions by simulating the training process in a noisy environment. Through this modeling, QAS can automatically identify which architectures are most robust under specific noise conditions, thereby ensuring that VQA performance is not excessively affected by noise.

In each round of optimization in quantum architecture search, MicroAlgo QAS not only considers changes in architecture design but also incorporates classical optimization algorithms such as gradient descent to ensure that the selected architecture can be efficiently trained for a given learning task. Through multiple iterations, QAS gradually converges to a quantum circuit architecture that both enhances expressive power and effectively mitigates the impact of noise. Furthermore, plateau phenomenon is another major challenge in VQA training. During the training process, optimization may encounter "barren plateau" regions, leading to local optima that make further improvements difficult. MicroAlgo QAS, through designing appropriate architectures and optimization strategies, can effectively avoid getting trapped in such barren plateaus, thereby improving the trainability and global optimization capability of VQA.

MicroAlgo QAS, by optimizing quantum circuit architectures, can significantly enhance the robustness of VQA in various noisy environments. By automatically searching for suitable circuit designs, QAS avoids the manual selection of unsuitable architectures, thereby enabling VQA to operate more effectively on actual quantum computers.

The optimization of quantum circuits is not merely about reducing the number of quantum gates; it is more about finding an architecture that can converge quickly and avoid getting trapped in local optima. Through intelligent search mechanisms and noise modeling, QAS enables VQA to complete training in a shorter time and ultimately find the global optimal solution.

Another advantage of MicroAlgo QAS is its broad adaptability. Whether used for quantum machine learning, quantum optimization problems, or quantum simulation tasks, QAS can adjust circuit architectures based on the requirements of different tasks, providing customized solutions. This makes QAS a highly flexible and practical tool in the field of quantum computing. MicroAlgo QAS is not only capable of running on current quantum devices but also possesses strong scalability. By optimizing circuit architectures, QAS can achieve more efficient operation on resource-constrained quantum computers, thereby making quantum computing more practical.

In multiple experimental validations, QAS has significantly outperformed traditional VQA approaches with manually designed circuit architectures. In standard quantum machine learning tasks, our QAS method has achieved remarkable results in reducing noise impact, improving training convergence speed, and mitigating the plateau effect. Compared to traditional methods, QAS has improved training speed by over 40% and enhanced robustness in noisy environments by 30%. Furthermore, in quantum optimization problems, QAS has similarly demonstrated powerful performance.

The launch of MicroAlgo's QAS technology marks a significant advancement in the application of Variational Quantum Algorithms (VQA). Through automated quantum circuit architecture search, QAS not only addresses issues such as noise, training efficiency, and the plateau effect but also significantly enhances the performance of VQA on real quantum computers. As quantum computing hardware continues to advance, QAS will become one of the core technologies in quantum algorithm development.

In the future, QAS can be applied not only to multiple fields such as quantum machine learning, quantum optimization, and quantum chemistry but also integrated with other advanced quantum computing technologies, such as quantum error correction and quantum communication, further promoting the popularization and application of quantum computing. We look forward to MicroAlgo's QAS laying a solid foundation for the commercial application of quantum computing, bringing more efficient and precise quantum solutions to various industries.

About MicroAlgo Inc.
MicroAlgo Inc. (the "MicroAlgo"), a Cayman Islands exempted company, is dedicated to the development and application of bespoke central processing algorithms. MicroAlgo provides comprehensive solutions to customers by integrating central processing algorithms with software or hardware, or both, thereby helping them to increase the number of customers, improve end-user satisfaction, achieve direct cost savings, reduce power consumption, and achieve technical goals. The range of MicroAlgo's services includes algorithm optimization, accelerating computing power without the need for hardware upgrades, lightweight data processing, and data intelligence services. MicroAlgo's ability to efficiently deliver software and hardware optimization to customers through bespoke central processing algorithms serves as a driving force for MicroAlgo's long-term development.

Forward-Looking Statements
This press release contains statements that may constitute "forward-looking statements." Forward-looking statements are subject to numerous conditions, many of which are beyond the control of MicroAlgo, including those set forth in the Risk Factors section of MicroAlgo's periodic reports on Forms 10-K and 8-K filed with the SEC. Copies are available on the SEC's website, www.sec.gov. Words such as "expect," "estimate," "project," "budget," "forecast," "anticipate," "intend," "plan," "may," "will," "could," "should," "believes," "predicts," "potential," "continue," and similar expressions are intended to identify such forward-looking statements. These forward-looking statements include, without limitation, MicroAlgo's expectations with respect to future performance and anticipated financial impacts of the business transaction.

MicroAlgo undertakes no obligation to update these statements for revisions or changes after the date of this release, except as may be required by law.

SOURCE MicroAlgo Inc.