Narrow Results

As the operational landscape of enterprises grows increasingly complex and volatile, the significance of modularization in economic management has become even more pronounced. By segmenting the management system into distinct yet interdependent modules, enterprises are better equipped to adapt swiftly to market fluctuations, enabling the efficient allocation of resources and the enhancement of management efficacy. Enterprise risk management, a pivotal component of modular management, faces unprecedented challenges, with traditional risk assessment methodologies often failing to meet the stringent demands for precision and real-time responsiveness. To overcome these challenges, this paper proposes a novel GT-DQN framework, integrating Graph Neural Networks (GNNs), transformer, and Deep Q-Network (DQN) algorithms to facilitate risk assessment within enterprise economic management. The framework undertakes comprehensive modeling of enterprise financial data, market transaction records, macroeconomic indicators, and supply chain relationships via GNN, while the transformer captures dynamic shifts in time series data. Ultimately, DQN optimizes risk decision-making strategies within an evolving economic environment, thereby enhancing the accuracy and stability of risk assessments. Experimental results demonstrate that the GT-DQN framework developed in this study achieves a recognition accuracy of 90% on public datasets across three tiers of enterprise risk — high, medium, and low — providing a robust technical foundation for future risk prediction and analysis in the modular management of enterprise economies.

In order to solve the problem of environment generalization in continuous state space, an obstacle avoidance method based on region location is proposed. The method is divided into three steps: (1) Using Region Proposal Network (RPN) to localize the obstacle area; (2). The environment map is established by the regional position mapping relation; and (3). The Deep Q-Learning Network (DQN) is used to realize collision detection of the robot, then pixel collision detection module is introduced and finally the pixel collision simulation distance sensor is combined to obtain the distance between the robot and obstacle and whether the collision or not. In this paper, the experiments were carried out in static obstacle environment and in dynamic and static obstacle environment for robot obstacle avoidance tasks. Experimental results show that the problem of environment generalization can be effectively solved by introducing pixel collision detection in the process of robot obstacle avoidance, and the network model trained in a dynamic environment has some generalization ability.

Reasonable allocation of resources is an important guarantee for efficient support of power business in edge IoT agents. Facing the above problems of the current power Internet of Things, this paper proposes a resource optimization allocation method based on deep Q-learning. This method first comprehensively considers the communication performance and network security. Involving indicators such as latency and service satisfaction, a complete and reliable mathematical model of the edge Internet of Things proxy network is constructed to achieve efficient and reliable modeling of the power Internet of Things (pIoT), aiming to better fit the practical interaction needs for efficient and secure communication. The Q-learning network model is optimized, and the method combining Reinforcement learning and deep learning is used to solve the model. Used by this network, the optimization and improvement of the deep network model is realized, so that the status, action and other parameters of the network model can be solved in a timely manner, so as to better support the reliable and efficient information interaction of the communication network. The test results prove that the delay of the proposed method can be maintained within 12ms in more complex scenarios, and the interaction success rate reaches 0.975, confirming that the proposed method can provide good information interaction guarantee services.

The prediction of price trends in the stock market has always been a hot research topic in the financial field. However, due to the high instability and volatility of stock prices, it is very difficult to accurately predict stock trends. How to remove the noise of stock data, extract effective features, and pursue maximum value returns has always been a challenge. This paper proposes a hybrid model (DWT-DQN) that combines discrete wavelet transform with deep reinforcement learning to improve the accuracy and return rate of stock predictions. First, the model captures price fluctuation information on different scales by performing discrete wavelet transformation on the difference between long- and short-term moving averages of stock prices, and well extracts the changing characteristics of stock price data in the time domain and frequency domain. Then the feature data are input into the built DQN network for model training. The network can select the optimal trading action based on market status and historical experience and returns. At the same time, during the data sampling process, an attention mechanism is introduced to allow the model to further learn in the direction of maximizing returns. Through testing and verification on SSEC, HSI, NDX and SPX data sets, experiments show that the hybrid model proposed in this paper has excellent performance in terms of accuracy and return rate.

Software Defined Networking (SDN) is a promising paradigm in the field of network technology. This paradigm suggests the separation between the control plane and the data plane which brings flexibility, efficiency and programmability to network resources.

SDN deployment in large scale networks raises many issues which can be overcame using a collaborative multi-controller approaches. Such approaches can resolve problems of routing optimization and network scalability. In large scale networks, such as SD-WAN, routing optimization consists of achieving a trade-off between per-flow QoS, the load balancing in each domain as well as the resource utilization in inter-domain links. Multi-Agent Reinforcement Learning paradigm(MARL) is one of the most popular solutions that can be used to optimize routing strategies in SD-WAN.

This paper proposes an efficient approach based on MARL which is able to ensure a load balancing among each network as well as optimized resource utilization of inter-domain links. This approach profits from our previous work, denoted SPFLR, and tries to balance the load of the whole network using Deep Q-Networks (DQN) algorithms. Simulation results show that the proposed solution performs better than parallel solutions such as BGP-based routing and random routing.

Reinforcement learning, as an effective method to solve complex sequential decision-making problems, plays an important role in areas such as intelligent decision-making and behavioral cognition. It is well known that the sample experience replay mechanism contributes to the development of current deep reinforcement learning by reusing past samples to improve the efficiency of samples. However, the existing priority experience replay mechanism changes the sample distribution in the sample set due to the higher sampling frequency assigned to a specific transition, and it cannot be applied to actor-critic and other on-policy reinforcement learning algorithm. To address this, we propose an adaptive factor based on TD-error, which further increases sample utilization by giving more attention weight to samples of larger TD-error, and embeds it flexibly into the original Deep Q Network and Advantage Actor-Critic algorithm to improve their performance. Then we carried out the performance evaluation for the proposed architecture in the context of CartPole-V1 and 6 environments of Atari game experiments, respectively, and the obtained results either on the conditions of fixed temperature or annealing temperature, when compared to those produced by the vanilla DQN and original A2C, highlight the advantages in cumulative rewards and climb speed of the improved algorithms.

Research on the acoustic micro-nano manipulation starts from the discovery of Chladni effect. Acoustic manipulation is expected to be applied to the culture of biological tissue and cell, micro nano element assembling, allocation of chemical raw materials and other micro nano scale fields. Meanwhile, acoustic manipulation shows the characteristic of contactless, biocompatibility, environmental compatibility and functional diversity. Whereas, the accuracy and intelligence of acoustic manipulation still have a big gap to be crossed. Very recently, the method of the deep reinforcement learning is hotly discussed, which provides a new idea for micro nano manipulation. In this paper, the Deep Q Network(DQN) algorithm is considered to improve the efficiency and intelligence in the process of acoustic manipulation. As a demonstration, linear motion tasks based on acoustic wave are trained and displayed. Consequently, the accurate acoustic frequency sequence can be obtained to direct the actual process of acoustic manipulation.