BEIJING, May 21, 2026 /PRNewswire/ — WiMi Hologram Cloud Inc. (NASDAQ: WiMi) (“WiMi” or the “Company”), a leading global Hologram Augmented Reality (“AR”) Technology provider, is researching quantum computing optimization based on multi-objective deep reinforcement learning. The core of this innovative solution lies in breaking the limitations of traditional single-objective optimization and constructing a global optimization framework that takes into account multi-dimensional constraints.
By using the single-process quantum control optimization results as the truncation threshold and reward function migration strategy for multi-objective optimization, effective reuse of optimization knowledge is achieved. This not only avoids redundant computation during the multi-objective optimization process but also improves the model’s convergence speed. At the same time, by designing a multi-objective reward function that comprehensively considers various key indicators in the quantum control process, synergistic optimization of multiple factors such as quantum gate fidelity, operational efficiency, noise suppression, and energy consumption control is realized, ultimately obtaining a globally optimal control solution rather than a locally optimal solution targeting only a single error metric, effectively improving the control precision and robustness of the quantum system. WiMi’s multi-objective deep reinforcement learning method, through deep learning and modeling of the dynamic characteristics of qubits, can adapt in real time to the dynamic changes of quantum systems, automatically adjust control strategies, and effectively suppress the impact of environmental noise and crosstalk effects.
The control of quantum systems essentially involves precisely regulating external physical fields to enable qubits to complete a series of processes such as state preparation, quantum gate operations, and state readout according to preset logic. Its core challenge lies in the openness and complexity of quantum systems—qubits are susceptible to environmental noise, crosstalk effects, decoherence, and other factors. Moreover, in multi-process quantum control, there exist multiple mutually constraining optimization objectives. Traditional control methods struggle to achieve global optimality. Traditional quantum control strategies are mostly based on model-driven optimization algorithms that rely on precise mathematical modeling of quantum systems. However, the dynamic characteristics of actual quantum systems are complex and easily affected by external interference, leading to deviations between the model and the actual system, which in turn affects control precision. At the same time, traditional methods mostly optimize for a single control objective, making them prone to falling into local optimal solutions. They cannot balance multi-dimensional requirements such as quantum gate fidelity, operation speed, and energy consumption control, making it difficult to adapt to the control scenarios of large-scale quantum systems.
The rapid iteration of machine learning technology has provided a completely new approach to solving the challenges of quantum control. Its powerful data-driven learning capability and adaptive optimization characteristics can effectively adapt to the complexity and uncertainty of quantum systems. Among them, reinforcement learning, as an important branch of machine learning, breaks through the dependence of traditional optimization algorithms on complete parameter sets. Through real-time interaction between the agent and the environment, it dynamically adjusts control strategies during the trial-and-error process to achieve gradual convergence of optimization objectives. This closed-loop mechanism of interaction-feedback-iteration highly aligns with the real-time control requirements of quantum systems, providing core technical support for the optimization of control strategies in quantum computing.
Quantum computing, as the core development direction of next-generation information technology, cannot achieve its practical application process without continuous breakthroughs in core technologies. In the future, WiMi will continue to focus on the forefront of quantum technology, taking technological innovation as the core driving force, deeply cultivating the interdisciplinary fields of quantum control, quantum algorithms, and artificial intelligence, continuously breaking through technical bottlenecks, promoting the development of quantum computing technology, and assisting various industries in achieving transformation and upgrading with the help of quantum 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.
