Multi-scenario and multi-task recommendation systems efficiently facilitate knowledge transfer across different scenarios and tasks. However, many existing approaches inadequately incorporate personalized information across users and scenarios. Moreover, the conversion rate (CVR) task in multi-task learning often encounters challenges like sample selection bias, resulting from systematic differences between the training and inference sample spaces, and data sparsity due to infrequent clicks. To address these issues, we propose Adaptive Entire-space Multi-scenario Multi-task Transfer Learning model (AEM$^{2}$TL) with four key modules: 1) Scenario-CGC (Scenario-Customized Gate Control), 2) Task-CGC (Task-Customized Gate Control), 3) Personalized Gating Network, and 4) Entire-space Supervised Multi-Task Module. AEM$^{2}$TL employs a multi-gate mechanism to effectively integrate shared and specific information across scenarios and tasks, enhancing prediction adaptability. To further improve task-specific personalization, it incorporates personalized prior features and applies a gating mechanism that dynamically scales the top-layer neural units. A novel post-impression behavior decomposition technique is designed to leverage all impression samples across the entire space, mitigating sample selection bias and data sparsity. Furthermore, an adaptive weighting mechanism dynamically allocates attention to tasks based on their relative importance, ensuring optimal task prioritization. Extensive experiments on one industrial and two real-world public datasets indicate the superiority of AEM$^{2}$TL over state-of-the-art methods.

Traffic flow forecasting is essential and challenging to intelligent city management and public safety. In this paper, we attempt to use a pure self-attention method in traffic flow forecasting. However, when dealing with input sequences, including large-scale regions’ historical records, it is difficult for the self-attention mechanism to focus on the most relevant ones for forecasting. To address this issue, we design a progressive space-time self-attention mechanism named ProSTformer, which can reduce self-attention computation times from thousands to tens. Our design is based on two pieces of prior knowledge in the traffic flow forecasting literature: (i) spatiotemporal dependencies can be factorized into spatial and temporal dependencies; (ii) adjacent regions have more influences than distant regions, and temporal characteristics of closeness, period and trend are more important than crossed relations between them. Our ProSTformer has two characteristics. First, each block in ProSTformer highlights the unique dependencies, ProSTformer progressively focuses on spatial dependencies from local to global regions, on temporal dependencies from closeness, period and trend to crossed relations between them, and on external dependencies such as weather conditions, temperature and day-of-week. Second, we use the Tensor Rearranging technique to force the model to compute self-attention only to adjacent regions and to the unique temporal characteristic. Then, we use the Patch Merging technique to greatly reduce self-attention computation times to distant regions and crossed temporal relations. We evaluate ProSTformer on two traffic datasets and find that it performs better than sixteen baseline models. The code is available at https://github.com/yanxiao1930/ProSTformer_code/tree/main.

Nonparallel support vector machine (NPSVM) is a novel nonparallel classifier for binary classification with large amounts of theoretical and practical advantages. How to train NPSVM efficiently on data with huge dimensions has not been studied. Recently, a variety of minwise hashing algorithms such as b-bit minwise hashing, connected bit minwise hashing and f-fractional bit minwise hashing are effectively applied to obtain a compact representation of the original data. However, they still have serious defects that the generation of k random permutations is time-consuming and the processing of the original dataset damages data structure. Fortunately, a simple and effective solution called one permutation hashing appears to avoid the disadvantages of the expensive preprocessing cost and the destruction of the original dataset. In this paper, we combine one permutation hashing scheme with linear NPSVM to speed up the training and testing phases for classification on large-scale and high dimensional datasets. Both theoretical analyses and experiments demonstrate that our algorithm achieves massive advantages in accuracy, efficiency and energy-consumption.

Multiview learning (MVL), by exploiting the complementary information among multiple feature sets, can improve the performance of many existing learning tasks. Support vector machine (SVM)-based models have been frequently used for MVL. A typical SVM-based MVL model is SVM-2K, which extends SVM for MVL by using the distance minimization version of kernel canonical correlation analysis. However, SVM-2K cannot fully unleash the power of the complementary information among different feature views. Recently, a framework of learning using privileged information (LUPI) has been proposed to model data with complementary information. Motivated by LUPI, we propose a new multiview privileged SVM model, multi-view privileged SVM model (PSVM-2V), for MVL. This brings a new perspective that extends LUPI to MVL. The optimization of PSVM-2V can be solved by the classical quadratic programming solver. We theoretically analyze the performance of PSVM-2V from the viewpoints of the consensus principle, the generalization error bound, and the SVM-2K learning model. Experimental results on 95 binary data sets demonstrate the effectiveness of the proposed method.

In this paper, we propose a new deep metric learning approach, called MVDML, for multi-view image classification. Multi-view features can provide more information than single view, however, it is a challenge to exploit and fuse the complementary information from multiple views. Multiple deep neural networks are constructed, each corresponds to a view, to extract nonlinear information from images. The nonlinear transformation is an improvement on linear transformation of metric learning. All the original images will be transformed into a lower-dimensional space. In each new space, the difference between intra-class distance and inter-class distance is maximized. To extract information from different views as much as possible, the difference between different views of the same image is minimized. The numerical experiments verify that our model can obtain competitive performance in image classification and runs faster than the baseline methods.

As a kind of popular problem in machine learning, multi-instance task has been researched by means of many classical methods, such as kNN, SVM, etc. For kNN classification, its performance on traditional task can be boosted by metric learning, which seeks for a data-dependent metric to make similar examples closer and separate dissimilar examples by a margin. It is a challenge to define distance between bags in multi-instance problem, let alone learning appropriate metric for the problem. In this paper, we propose a new approach for multi-instance classification, with the idea of metric learning embedded. A new kind of distance is used to measure the similarity between bags. To weaken redundant information from bags and reduce computation complexity, k-means method is implemented to get collapsed bags by replacing each instance with its corresponding cluster centroid. The aim of metric learning is to expand inter-class bag distance and shrink intra-class bag distance, leading to the construction of an optimization problem with maximal relative distance. Kernel function can be introduced into the model to extract nonlinear information from the inputs. Gradient descent is utilized to solve the problem effectively. Numerical experiments on both artificial datasets and benchmark datasets demonstrated that the method can obtain competitive performance comparative to kNN and the state-of-the-art method in multi-instance classification.

In recent years, nonparallel support vector machine (NPSVM) is proposed as a nonparallel hyperplane classifier with superior performance than standard SVM and existing nonparallel classifiers such as the twin support vector machine (TWSVM). With the perfect theoretical underpinnings and great practical success, NPSVM has been used to dealing with the classification tasks on different scales. Tackling large-scale classification problem is a challenge yet significant work. Although large-scale linear NPSVM model has already been efficiently solved by the dual coordinate descent (DCD) algorithm or alternating direction method of multipliers (ADMM), we present a new strategy to solve the primal form of linear NPSVM different from existing work in this paper. Our algorithm is designed in the framework of the stochastic gradient descent (SGD), which is well suited to large-scale problem. Experiments are conducted on five large-scale data sets to confirm the effectiveness of our method.