Deep reinforcement learning (DRL) has superior autonomous decision-making capabilities, combining deep learning and reinforcement learning (RL). Unlike DRL employs deep neural networks (DNNs), broad RL (BRL) adopts the broad learning system (BLS) that is established with flat networks to generate the strategy. This article proposes the multiagent adaptive broad-DRL (ABDRL) approach for traffic light control (TLC), which combines the broad network with the deep network structure. Specifically, the structure of ABDRL first expands in the form of flatted broad networks. Then, the feature representation module that contains DNNs is employed to extract the critical traffic information. In addition, experiences sampled randomly by the experience replay mechanism cannot reflect the current training status of the agent effectively. In order to alleviate the impacts caused by random sampling, the forgetful experience mechanism (FEM) is incorporated into ABDRL. The FEM enables the agent to discriminate the importance of experiences stored in the experience reply buffer to improve robustness and adaptability. We validate the effectiveness of ABDRL in TLC, and the results illustrate the optimality and robustness of ABDRL over the state-of-the-art multiagent DRL (MADRL) algorithms.

Despite significant advancements in Multi-Agent Deep Reinforcement Learning (MADRL) approaches for Traffic Light Control (TLC), effectively coordinating agents in diverse traffic environments remains a challenge. Studies in MADRL for TLC often focus on repeatedly constructing the same intersection models with sparse experience. However, real road networks comprise Multi-Type of Intersections (MTIs) rather than being limited to intersections with four directions. In the scenario with MTIs, each type of intersection exhibits a distinctive topology structure and phase set, leading to disparities in the spaces of state and action. This paper introduces Adaptive Multi-agent Deep Mixed Reinforcement Learning (AMDMRL) for addressing tasks with multiple types of intersections in TLC. AMDMRL adopts a two-level hierarchy, where high-level proxies guide lowlevel agents in decision-making and updating. All proxies are updated by value decomposition to obtain the globally optimal policy. Moreover, the AMDMRL approach incorporates a mixed cooperative mechanism to enhance cooperation among agents, which adopts a mixed encoder to aggregate the information from correlated agents.We conduct comparative experiments involving four traditional and four DRL-based approaches, utilizing three training and four testing datasets. The results indicate that the AMDMRL approach achieves average reductions of 41% than traditional approaches, and 16% compared to DRL-based approaches in traveling time on three training datasets. During testing, the AMDMRL approach exhibits a 37% improvement in reward compared to the MADRL-based approaches.

Knowledge graphs are extensively utilized in diverse fields such as search engines, recommendation systems, and dialogue systems, and knowledge graph reasoning plays an important role in the aforementioned domains. Graph neural networks demonstrate the capability to effectively capture and process the graph structure inherent in knowledge graphs, leveraging the relationships between nodes and edges to enable efficient reasoning. Current research on graph neural networks relies on predefined propagation paths. The models based on predefined propagation paths overlook the correlation between entities and query relations, limiting adaptability and scalability. In this paper, we propose an adaptive graph neural network with an incremental learning mechanism to search the adaptive propagation path in order to retain promising targets. In detail, the incremental learning mechanism is able to filter out unrelated entities in the propagation path by incorporating the node sampling technique to increase the learning efficiency of the model. In addition, the incremental learning mechanism can select semantically related entities, which promotes the capture of meaningful connections among different nodes in the knowledge graph. At the same time, we apply the parameter initialization module to accelerate the convergence and improve the robustness of the model. Experimental results on benchmark datasets demonstrate that the adaptive graph neural network with the incremental learning mechanism has excellent semantic awareness ability, which surpasses several excellent knowledge graph reasoning models.