Space-Air-Ground Integrated Networks (SAGINs) are a cornerstone of future 6 G systems, promising ubiquitous connectivity by amalgamating satellite, aerial, and terrestrial communication resources. However, these networks traditionally operate in isolated silos, leading to inefficient resource utilization and limited adaptability. To address these challenges, this paper proposes a collaborative access framework that allows cross-layer communication, enabling user equipment from one domain to access network infrastructure in another. The key innovations of this work lie in three aspects: first, it breaks the fixed hierarchy of traditional SAGINs and integrates access selection into the optimization problem, allowing dynamic adjustment of useraccess point associations; second, it establishes a holistic mathematical model that simultaneously considers access selection, frequency resource block allocation, and power control, capturing the complex interference interactions between different layers; third, it provides a theoretical benchmark for evaluating the performance of practical algorithms by formulating the problem as a unified MINLP model.

We present a novel deep reinforcement learning framework for optimizing telecom product portfolios. Our proposed Telecom Product Portfolio Optimization (TPPO) algorithm formulates portfolio consolidation as a sequential decision-making problem, integrating multi-dimensional product feature representation and a specially designed multi-objective reward function. The framework employs deep Q-learning to balance portfolio reduction with revenue preservation and user experience. Experimental results demonstrate that TPPO achieves a 33.8% portfolio reduction while increasing revenue by 5.47%, significantly outperforming traditional rule-based and clustering approaches. The framework demonstrates strong practical value for telecom operators seeking to streamline their product portfolios while enhancing business performance.

In content-based image retrieval (CBIR), one of the most challenging and ambiguous tasks are to correctly understand the human query intention and measure its semantic relevance with images in the database. Due to the impressive capability of visual saliency in predicting human visual attention that is closely related to the query intention, this paper attempts to explicitly discover the essential effect of visual saliency in CBIR via qualitative and quantitative experiments. Toward this end, we first generate the fixation density maps of images from a widely used CBIR dataset by using an eye-tracking apparatus. These ground-truth saliency maps are then used to measure the influence of visual saliency to the task of CBIR by exploring several probable ways of incorporating such saliency cues into the retrieval process. We find that visual saliency is indeed beneficial to the CBIR task, and the best saliency involving scheme is possibly different for different image retrieval models. Inspired by the findings, this paper presents two-stream attentive CNNs with saliency embedded inside for CBIR. The proposed network has two streams that simultaneously handle two tasks. The main stream focuses on extracting discriminative visual features that are tightly related to semantic attributes. Meanwhile, the auxiliary stream aims to facilitate the main stream by redirecting the feature extraction to the salient image content that human may pay attention to. By fusing these two streams into the Main and Auxiliary CNNs (MAC), image similarity can be computed as the human being does by reserving conspicuous content and suppressing irrelevant regions. Extensive experiments show that the proposed model achieves impressive performance in image retrieval on four public datasets.

In content-based image retrieval, the most challenging (and ambiguous) part is to define the similarity between images. For the human-being, such similarity can be defined with respect to where they pay attention to and what semantic attributes they understand. Inspired by this fact, this paper presents two-stream attentive CNNs for image retrieval. As the human-being does, the proposed network has two streams that simultaneously handle two tasks. The Main stream focuses on extracting discriminative visual features that are tightly correlated with semantic attributes. Meanwhile, the Auxiliary stream aims to facilitate the main stream by redirecting the feature extraction operation mainly to the image content that human may pay attention to. By fusing these two streams into the Main and Auxiliary CNNs (MAC), image similarity can be computed as the human-being does by reserving the conspicuous content and suppressing the irrelevant regions. Extensive experiments show that the proposed model achieves impressive performance in image retrieval on four public datasets.