Recent studies employing deep learning to solve the traveling salesman problem (TSP) have mainly focused on learning construction heuristics. Such methods can improve TSP solutions, but still depend on additional programs. However, methods that focus on learning improvement heuristics to iteratively refine solutions remain insufficient. Traditional improvement heuristics are guided by a manually designed search strategy and may only achieve limited improvements. This paper proposes a novel framework for learning improvement heuristics, which automatically discovers better improvement policies for heuristics to iteratively solve the TSP. Our framework first designs a new architecture based on a transformer model to make the policy network parameterized, which introduces an action-dropout layer to prevent action selection from overfitting. It then proposes a deep reinforcement learning approach integrating a simulated annealing mechanism (named RL-SA) to learn the pairwise selected policy, aiming to improve the 2-opt algorithm's performance. The RL-SA leverages the whale optimization algorithm to generate initial solutions for better sampling efficiency and uses the Gaussian perturbation strategy to tackle the sparse reward problem of reinforcement learning. The experiment results show that the proposed approach is significantly superior to the state-of-the-art learning-based methods, and further reduces the gap between learning-based methods and highly optimized solvers in the benchmark datasets. Moreover, our pre-trained model M can be applied to guide the SA algorithm (named M-SA (ours)), which performs better than existing deep models in small-, medium-, and large-scale TSPLIB datasets. Additionally, the M-SA (ours) achieves excellent generalization performance in a real-world dataset on global liner shipping routes, with the optimization percentages in distance reduction ranging from 3.52% to 17.99%.

Livestock transportation is a key factor that contributes to the spatial spread of brucellosis. To analyze the impact of sheep transportation on brucellosis transmission, we develop a human-sheep coupled brucellosis model within a metapopulation network framework. Theoretically, we examine the positively invariant set, the basic reproduction number, the existence, uniqueness, and stability of disease-free equilibrium and the existence of the endemic equilibrium of the model. For practical application, using Heilongjiang province as a case study, we simulate brucellosis transmission across 12 cities based on data using three network types: the BA network, the ER network, and homogeneous mixing network. The simulation results indicate that the network's average degree plays a role in the spread of brucellosis. For BA and ER networks, the basic reproduction number and cumulative incidence of brucellosis stabilize when the network's average degree reaches 4 or 5. In contrast, sheep transport in a homogeneous mixing network accelerates the cross-regional spread of brucellosis, whereas transportation in a BA network helps to control it effectively. Furthermore, the findings suggest that the movement of sheep is not always detrimental to controlling the spread of brucellosis. For cities with smaller sheep populations, such as Shuangyashan and Qitaihe, increasing the transport of sheep outward amplifies the spatial spread of the disease. In contrast, in cities with larger sheep populations, such as Qiqihar, Daqing, and Suihua, moderate sheep outflow can help reduce the spread. In addition, cities with large livestock populations play a dominant role in the overall transmission dynamics, underscoring the need for stricter supervision in these areas.

Essential proteins are crucial for biological processes and can be identified through both experimental and computational methods. While experimental approaches are highly accurate, they often demand extensive time and resources. To address these challenges, we present a computational ensemble learning framework designed to identify essential proteins more efficiently. Our method begins by using node2vec to transform proteins in the protein-protein interaction (PPI) network into continuous, low-dimensional vectors. We also extract a range of features from protein sequences, including graph-theory-based, information-based, compositional, and physiochemical attributes. Additionally, we leverage deep learning techniques to analyze high-dimensional position-specific scoring matrices (PSSMs) and capture evolutionary information. We then combine these features for classification using various machine learning algorithms. To enhance performance, we integrate the outputs of these algorithms through ensemble methods such as voting, weighted averaging, and stacking. This approach effectively addresses data imbalances and improves both robustness and accuracy. Our ensemble learning framework achieves an AUC of 0.960 and an accuracy of 0.9252, outperforming other computational methods. These results demonstrate the effectiveness of our approach in accurately identifying essential proteins and highlight its superior feature extraction capabilities.