Ensemble learning has achieved remarkable success in machine learning, but its reliance on numerous base learners limits its application in resource-constrained environments. This paper introduces an innovative "Margin-Maximizing Fine-Grained Ensemble Method" that achieves performance surpassing large-scale ensembles by meticulously optimizing a small number of learners and enhancing generalization capability. We propose a novel learnable confidence matrix, quantifying each classifier’s confidence for each category, precisely capturing category-specific advantages of individual learners. Furthermore, we design a margin-based loss function, constructing a smooth and partially convex objective using the logsumexp technique. This approach improves optimization, eases convergence, and enables adaptive confidence allocation. Finally, we prove that the loss function is Lipschitz continuous, based on which we develop an efficient gradient optimization algorithm that simultaneously maximizes margins and dynamically adjusts learner weights. Extensive experiments demonstrate that our method outperforms traditional random forests using only one-tenth of the base learners and other state-of-the-art ensemble methods.

Fusing heterogeneous information remains a persistent challenge in modern data analysis. While significant progress has been made, existing approaches often fail to account for the inherent heterogeneity of object patterns across different semantic spaces. To address this limitation, we propose the **Cooperation of Experts (CoE)** framework, which encodes multi-typed information into unified heterogeneous multiplex networks. By transcending modality and connection differences, CoE provides a powerful and flexible model for capturing the intricate structures of real-world complex data. In our framework, dedicated encoders act as domain-specific experts, each specializing in learning distinct relational patterns in specific semantic spaces. To enhance robustness and extract complementary knowledge, these experts collaborate through a novel **large margin** mechanism supported by a tailored optimization strategy. Rigorous theoretical analyses guarantee the framework’s feasibility and stability, while extensive experiments across diverse benchmarks demonstrate its superior performance and broad applicability.

Recent advances in unsupervised deep graph clustering have been significantly promoted by contrastive learning. Despite the strides, most graph contrastive learning models face challenges: 1) graph augmentation is used to improve learning diversity, but commonly used random augmentation methods may destroy inherent semantics and cause noise; 2) the fixed positive and negative sample selection strategy ignores the difficulty distribution of samples when deal with complex real data, thereby impeding the model’s capability to capture fine-grained patterns and trapping the model in sub-optimal for clustering. To reduce these problems, we propose the Clustering-guided Curriculum Graph contrastive Learning (CurGL) framework. CurGL uses clustering entropy as the guidance of the following graph augmentation and contrastive learning. Specifically, according to the clustering entropy, the intra-class edges and important features are emphasized in augmentation. Then, a multi-task curriculum learning scheme is proposed, which employs the clustering guidance to shift the focus from the discrimination task to the clustering task. In this way, the sample selection strategy of contrastive learning can be adjusted adaptively from early to late stage, which enhances the model's flexibility for complex data structure. Experimental results demonstrate that CurGL has achieved excellent performance compared to state-of-the-art competitors.

Ensemble clustering based on co-association matrices integrates multiple connective matrices from base clusterings to achieve superior results. However, these methods primarily focus on inter-sample relationships, neglecting variations across different base clusterings, potentially introducing noise. Additionally, they overlook interactions between samples and base clusterings, which are crucial for extracting common information and avoiding post-processing steps that may cause information loss and instability in clustering results. To address these issues, we propose the Tensorized Graph Learning for Spectral Ensemble Clustering (TGLSEC) model. TGLSEC stacks all connective matrices into a third-order tensor, employs Fast Fourier Transform (FFT) for encoding, and elucidates inter-relations in the frequency domain. By minimizing the tensor Schatten p-norm, TGLSEC extracts common information in the low-rank space, eliminating noise and improving the quality of the common shared graph. Incorporating Laplacian rank constraints, TGLSEC learns a common shared graph with c-connected components, directly representing the clustering structure and avoiding post-processing steps, leading to more stable clustering results. To enhance computational efficiency for large-scale datasets, TGLSEC has been expanded into a bipartite-graph-based model, TGLSEC-BG, reducing complexity and computational time. Extensive experiments on real-world datasets demonstrate that TGLSEC and TGLSEC-BG exhibit superior clustering performance and robustness to noise.

Speed prediction is a crucial yet complicated task for intelligent transportation systems. The challenge derives from the complex spatiotemporal dependencies of traffic parameters. In the past few years, deep neural networks have achieved the best traffic speed prediction performance. However, most models depend on short-term input sequences to predict short/long-term traffic speed (e.g., predicting speed for the next hour using data from the past hour). These models fail to consider the daily and weekly periodic behavior of traffic. Another problem posed by neural networks is the lack of interpretability as they often operate as “black boxes”. In this paper, an attention-based multi-encoder-decoder (Att-MED) model is proposed to predict traffic speed. The model uses convolutional-LSTMs to capture the spatiotemporal relationship of multiple input sequences, namely short-term, daily and weekly traffic patterns. The model also employs an LSTM to model the output predictions sequentially. Furthermore, attention mechanism is used to weigh the contribution of each traffic sequence towards the output predictions. The proposed network architecture, when trained end-to-end, results in a superior prediction accuracy compared to baseline models. In addition to contributing towards performance, the attention mechanism creates weight values, which when visualized, provide insights into the decision-making process of the neural network, and consequently produce explainable outputs. Att-MED’s extracted attention weights highlight the contribution of daily and weekly periodic input towards speed prediction.

Proactive traffic safety management systems can reduce crashes by identifying crash precursors, evaluating real-time crash risks, and implementing suitable interventions. The basic prerequisite for developing such a system is to propose a reliable crash risk evaluation model that takes real-time traffic flow data as input. Previous studies have primarily focused on real-time crash prediction using some statistical or machine-learning methods. However, further quantitative evaluation and classification of crash risks have been ignored. In this study, we conduct a systematic crash risk evaluation workflow, including crash risk prediction, crash risk quantification, and crash risk classification. Specifically, the crash risk prediction using an extended logit model is proposed, from which CAS, CSD, UAS, DAS, DTV are identified to be contributing factors of crash risks. Then a crash risk quantification model based on the parameter evaluation of the extended logit model is developed. The crash risks of urban expressways and their spatial-temporal evolution trends are quantified. Finally, the crash risks are classified into high crash risk level, moderate crash risk level, and low crash risk level by the k-means cluster algorithm. Then the threshold boundaries of different crash risk levels are determined. The research results provide a proactive guidance for traffic safety management of urban expressways.

In the context of pro-active traffic management, real-time crash risk evaluation is one of the most critical components. Signalized intersections are well-known high-risk locations because of the variety of traffic movements, modes, and their interactions. Unlike access-controlled freeways, the traffic flow at signalized intersections presents cyclical characteristics, which are temporally separated by traffic signals. Therefore, the data preparation for real-time crash risk prediction at signalized intersections should be based on the signal cycle rather than a predefined fixed time interval (e.g., 5 minutes). In this research, the corresponding signal cycles where crashes occurred were verified based on high-resolution event-based data (i.e., Automated Traffic Signal Performance Measures (ATSPM)). Six types of real-time cycle-level factors were considered, including traffic volume, signal timing, headway and occupancy, traffic variation between upstream and downstream detectors, shockwave characteristics, and weather. Two undersampling strategies, matched case-control and random undersampling, were utilized to develop conditional logistic and binary logistic models, respectively. Model results indicate that the random undersampling performs better than the matched case-control method. It was found that higher cycle volume, overall average flow ratio across lanes, arrivals on yellow ratio, traffic volatility across approach sections, as well as longer cycle length and lower green ratio could significantly increase the odds of crash occurrence at signalized intersections. Moreover, longer maximum queue length, bigger shockwave, and higher absolute queuing shockwave speed tend to increase the odds of crash occurrence.