While matrix factorization based method is a popular choice in recommendation, several recent efforts have shown that the inner product does not satisfy the triangle inequality, which may hurt the model's generalization ability. TransRec is a promising method to overcome this issue, which learns a distance metric to predict the strength of user-item interactions. Nevertheless, such method only uses the latest consumed item to model a user's short-term preference, which is insufficient for modeling fidelity. In this paper, we propose an Attentive Translation Model (ATM) to exploit high-order sequential information for next-item recommendation. Specifically, we construct a user-specific translation vector by accounting for multiple recent items, which encode more information about a user's short-term preference than the latest item. To aggregate multiple items into one representation, we devise a position-aware attention mechanism for ATM. Extensive experiments on multiple datasets show that ATM significantly outperforms several state-of-the-arts.

Modeling sequential behaviors is the core of sequential recommendation. As users visit items in chronological order, existing methods typically capture a user’s present interests from his/her past-to-present behaviors, i.e. making recommendations with only the unidirectional past information. This article argues that future information is another critical factor for the sequential recommendation. However, directly learning from future-to-present behaviors inevitably causes data leakage. Here, it is pointed out that future information can be learned from users’ collaborative behaviors. Toward this end, this article introduces sequential graphs to depict item transition relationships: where and how each item transits from and will transit to. This temporal evolution information is called the light cone in special and general relativity. Then, a bidirectional sequential graph convolutional network (BiSGCN) is proposed to learn item representations by encoding past and future light cones. Finally, a manifold translating embedding (MTE) method is proposed to model item transition patterns in Riemannian manifolds, which helps to better capture the geometric structures of light cones and item transition patterns. Experimental comparisons and ablation studies verify the outstanding performance of BiSGCN, the benefits of learning from the future, and the improvements of learning in Riemannian manifolds.

Generative adversarial nets (GANs) have enjoyed considerable success in computer vision and attracted much attention from recommender systems. However, due to the discrete nature of items, it is infeasible to graft GANs directly onto recommendation models. Although several methods have taken steps forward, their training processes are slow-convergent, time-consuming, or even unstable. This article proposes a novel framework named attentive adversarial collaborative filtering (AACF) and an efficient training strategy to improve GANs in recommender systems. There are two distinct novelties over previous work. First, AACF is a differentiable generative adversarial framework that introduces an attention mechanism and “virtual items” to bridge the gap between the generator and the discriminator. Owing to the intrinsic differentiability, AACF can be stably optimized with gradient descent methods. Second, the efficient training strategy substantially reduces computational complexity. It is capable of efficiently training and scaling up the AACF model to large datasets. Extensive experiments on various datasets demonstrate the effectiveness, fast convergence, stability, and scalability of AACF. Since our ideas are general in nature, they will open a path to stably and efficiently train GANs in the research areas with discrete data. The implementation code is available at https://github.com/zhongchuansun/AACF.

Remote sensing object detection (RSOD) is highly challenging due to large variations in object scales. Existing deep learning-based methods still face limitations in addressing this challenge. Specifically, reliance on stride convolutions during downsampling leads to the loss of object information, and insufficient context-aware modeling capability hampers full utilization of object information at different scales. To address these issues, this paper proposes a Haar wavelet-based Attention Network (HWANet). The model includes a Low-frequency Enhanced Downsampling Module (LEM), a Haar Frequency Domain Self-attention Module (HFDSA), and a Spatial Information Interaction Module (SIIM). Specifically, LEM employs the Haar wavelet transform to downsample feature maps and enhances low-frequency components, mitigating the loss of object information at different scales. The HFDSA module integrates Haar wavelet transform and explicit spatial priors, reducing computational complexity while enhancing the capture of image spatial structures. Meanwhile, the SIIM module facilitates interactions among information at different levels, enabling multi-level feature integration. Together, SIIM and HFDSA strengthen the model’s context-aware modeling capability, allowing full utilization of multi-scale information. Experimental results show that HWANet achieves 93.1% mAP50 on the NWPU VHR-10 dataset and 99.1% mAP50 on the SAR-Airport-1.0 dataset, with only 2.75M parameters, outperforming existing methods.