Chinese Standard Mahjong is a complex, multi-player game with incomplete information, characterized by a vast strategic space, hidden information, and high decision-making uncertainty. These factors pose significant challenges in training efficient Artificial Intelligence (AI) systems, particularly when computational resources are constrained, due to sparse rewards and delayed feedback. To address these challenges, we propose MJ_RM , an AI model based on Reinforcement Learning (RL) that integrates an improved Distributed Proximal Policy Optimization (DPPO) algorithm with a multi-dimensional, expert-driven, three-stage reward mechanism . This reward design strengthens the link between intermediate actions and final outcomes, accelerating convergence. The model architecture combines a Residual Residual Network (Res2Net) with a Long Short-Term Memory (LSTM) network, jointly capturing multi-scale spatial features and long-range temporal dependencies inherent in Mahjong gameplay. We explicitly define the optimization objectives, including policy loss, value loss, and entropy regularization, and outline key hyperparameters, such as the discount factor γ and smoothing coefficient λ , used in our experiments. Additionally, we provide a sensitivity analysis of reward parameters and ablations of architectural components to enhance reproducibility Experimental evaluations on the Botzone platform demonstrate that MJ_RM achieves strong performance under low-resource constraints and exhibits cross-game generalization to Gomoku and Dou Dizhu, showing its applicability beyond Mahjong-specific scenarios. We also discuss the model’s computational efficiency, limitations, and outline future directions, including multi-agent extensions.

Traffic flow prediction is a critical technology in intelligent transportation management systems, contributing to alleviating traffic congestion and optimizing resource allocation. However, the nonlinear and complex spatiotemporal characteristics of traffic flow data pose significant challenges to accurate prediction. To address these issues, this paper proposes a hybrid model, CL-KAN, which combines Convolutional Neural Networks, Long Short-Term Memory networks, and Kolmogorov-Arnold Networks. The model improves prediction performance and computational efficiency through spatial feature extraction, temporal sequence modeling, and nonlinear feature integration. Experimental results demonstrate that the CL-KAN model achieves prediction performance comparable to the Transformer model while significantly reducing the number of parameters and improving computational efficiency, making it particularly suitable for resource-constrained scenarios. Furthermore, compared to other models, CL-KAN exhibits superior performance in capturing complex spatiotemporal dependencies and dynamic changes. This study validates the efficiency and practicality of the CL-KAN model, providing a feasible solution for real-time prediction and complex scenario applications in intelligent transportation systems.

Tibetan Jiu chess, recognized as a national intangible cultural heritage, is a complex game comprising two distinct phases: the layout phase and the battle phase. Improving the performance of deep reinforcement learning (DRL) models for Tibetan Jiu chess is challenging, especially given the constraints of hardware resources. To address this, we propose a two-stage model called JFA, which incorporates hierarchical neural networks and knowledge-guided techniques. The model includes sub-models: strategic layout model (SLM) for the layout phase and hierarchical battle model (HBM) for the battle phase. Both sub-models use similar network structures and employ parallel Monte Carlo tree search (MCTS) methods for independent self-play training. HBM is structured as a hierarchical neural network, with the upper network selecting movement and jump capturing actions and the lower network handling square capturing actions. Human knowledge-based auxiliary agents are introduced to assist SLM and HBM, simulating the entire game and providing reward signals based on square capturing or victory outcomes. Additionally, within the HBM, we propose two human knowledge-based pruning methods that prune parallel MCTS and capture actions in the lower network. In the experiments against a layout model using the AlphaZero method, SLM achieves a 74% win rate, with the decision-making time being reduced to approximately 1/147 of the time required by the AlphaZero model. SLM also won the first place at the 2024 China National Computer Game Tournament. HBM achieves a 70% win rate when playing against other Tibetan Jiu chess models. When used together, SLM and HBM in JFA achieve an 81% win rate, comparable to the level of a human amateur 4-dan player. These results demonstrate that JFA effectively enhances artificial intelligence (AI) performance in Tibetan Jiu chess.

The rules of layout, battle and flying stages of Tibetan Jiu chess are different, which leads to different spatial complexity in each stage. This paper proposes a nested three-stage game algorithm to better enhance the level of AI playing. An alpha-beta pruning algorithm based on chess shapes is designed for the layout state. The battle stage is subdivided into three nested sub-stages. The first sub-stage adopts the alpha-beta pruning algorithm based on chess shape evaluation, the second sub-stage adopts the more complex chess shape evaluation, and the third-sub stage adopts the local MCTS algorithm based on chess shape. A chess shape evaluation algorithm is proposed for the flying stage. We developed a 14-way Tibetan Jiu chess AI, named Evergreen, using proposed algorithm. Evergreen won the first prize in the Tibetan Jiu chess competition at 2023 China University Computer Game Competition and Game Championships. Through comprehendsive evaluation, the AI developed using proposed algorithm has reached the level of two dan amateur player. The results of experiments and competitions between human players verified that the designed three-stage nested algorithm has high level of making action decision.

Tibetan Jiu Chess is a national intangible cultural heritage of China.Developing digital online chess platform is one of the important means to protect and inherit Jiu Chess. Based on Unity3D game engine, the author developed a 14-way Tibetan Jiu Chess online playing platform. The platform can realize the automatic play between Jiu Chess AI program and online game play for human players.

This article proposes a novel Mahjong game model, LsAc ∗‐MJ, designed to address challenges posed by data scarcity, difficulty in leveraging contextual information, and the computational resource‐intensive nature of self‐play zero‐shot learning. The model is applied to Japanese Mahjong for experiments. LsAc ∗‐MJ employs long short‐term memory (LSTM) neural networks, utilizing hidden nodes to store and propagate contextual historical information, thereby enhancing decision accuracy. Additionally, the paper introduces an optimized Advantage Actor‐Critic (A2C) algorithm incorporating an experience replay mechanism to enhance the model’s decision‐making capabilities and mitigate convergence difficulties arising from strong data correlations. Furthermore, the paper presents a two‐stage training approach for self‐play deep reinforcement learning models guided by expert knowledge, thereby improving training efficiency. Extensive ablation experiments and performance comparisons demonstrate that, in contrast to other typical deep reinforcement learning models on the RLcard platform, the LsAc ∗‐MJ model consumes lower computational and time resources, has higher training efficiency, faster average decision time, higher win‐rate, and stronger decision‐making ability.

The game of Tibetan Go faces the scarcity of expert knowledge and research literature. Therefore, we study the zero learning model of Tibetan Go under limited computing power resources and propose a novel scale-invariant U-Net style two-headed output lightweight network TibetanGoTinyNet. The lightweight convolutional neural networks and capsule structure are applied to the encoder and decoder of TibetanGoTinyNet to reduce computational burden and achieve better feature extraction results. Several autonomous self-attention mechanisms are integrated into TibetanGoTinyNet to capture the Tibetan Go board’s spatial and global information and select important channels. The training data are generated entirely from self-play games. TibetanGoTinyNet achieves 62%–78% winning rate against other four U-Net style models including Res-UNet, Res-UNet Attention, Ghost-UNet, and Ghost Capsule-UNet. It also achieves 75% winning rate in the ablation experiments on the attention mechanism with embedded positional information. The model saves about 33% of the training time with 45%–50% winning rate for different Monte-Carlo tree search (MCTS) simulation counts when migrated from 9 × 9 to 11 × 11 boards. Code for our model is available at https://github.com/paulzyy/TibetanGoTinyNet .

The rules of the two phases of Tibetan Jiu chess, layout and battle, are very different, and using the same UCT search algorithm globally will result in a large overhead of time and storage space in the search process, so a phased game algorithm for Tibetan Jiu chess is proposed, with different strategies designed for the layout and battle phases, respectively. First, the layout phase uses a combination of Gaussian distribution and fast online estimation to improve the UCT algorithm, thus generating the optimal action selection scheme. Second, in order to take full advantage of reinforcement learning and deep learning, a neural network model with residual network structure is used in the battle phase to guide the search of Monte Carlo trees, and the default strategy is improved by "pruning" in the expansion step to improve the quality of the expanded nodes. The dataset is generated by self-play and used to train the neural network model to obtain the optimal model. It is verified through experiments that the phased gaming algorithm proposed in this study effectively reduces the process of blindly exploring the board state during the layout and battle phases of the UCT search algorithm, and improves the quality of the layout and the self-learning efficiency of the neural network model.

The sparrow search algorithm (SSA) has received widespread attention as an emerging group intelligence algorithm. In this study, a QSAR (quantitative structure-activity relationship) prediction model for anti-ERa (Estrogen receptor alpha, an important target for the treatment of breast cancer) compounds and their bioactivity data was constructed based on a sparrow search algorithm optimized BP neural network combined with the selected relevant parameter indicators. Then the created model was used to predict the biological activity of the compounds. The results show that the created model not only has a certain self-learning function, but also improves the convergence speed and the accuracy of the prediction results compared with the BP neural network model optimized by genetic algorithm (which is more complex in coding and slower in solving, but has good solution accuracy). Further, it was demonstrated that the use of SSA-BP could improve the prediction of QSAR for anti-ERa compounds.

To tackle the problem of aquatic environment pollution, a vision-based autonomous underwater garbage cleaning robot has been developed in our laboratory. We propose a garbage detection method based on a modified YOLOv4, allowing high-speed and high-precision object detection. Specifically, the YOLOv4 algorithm is chosen as a basic neural network framework to perform object detection. With the purpose of further improvement on the detection accuracy, YOLOv4 is transformed into a four-scale detection method. To improve the detection speed, model pruning is applied to the new model. By virtue of the improved detection methods, the robot can collect garbage autonomously. The detection speed is up to 66.67 frames/s with a mean average precision (mAP) of 95.099%, and experimental results demonstrate that both the detection speed and the accuracy of the improved YOLOv4 are excellent.

Machine learning algorithms are widely applied in the fields such as Machine Vision, Natural Language Processing, Image Processing and Automatic Speech Recognition. Deep learning based on mass data has achieved success in many application, for example, Alpha Go. However, learning from small data remains a key challenge in machine learning. This paper introduces the basic ideas of the small data learning theory, present the main machine learning algorithms and analyze their major characteristics. The paper also summarizes the current research trends.