Unsupervised federated learning for cross-modal retrieval has received increasing attention in recent years as it can free the requirement for annotations and avoid uploading original clients’ data to servers. Most existing methods focus on how to learn better local models and their aggregation to overcome data distribution drift across clients. Unlike prior works, we propose to address the data distribution problem by generating synthetic data, which can benefit existing federated learning methods. Specifically, we train a WGAN generator with three newly designed loss constraints on each client to improve the quality of the generated data. We first compute cluster prototypes to address the problem of lack of labels. Then, a direct contrastive loss between generated image and text features, an indirect contrastive loss with reference to cluster prototypes, and a Jensen-Shannon Divergence (JSD) loss also with reference to cluster prototypes work together to constrain the WGAN. The locally trained generators and local prototypes are sent to the server to generate and filter synthetic data with consideration of data distribution across all clients. The filtered data are used to train the aggregated global retrieval model, which is later sent to clients. The final global model becomes robust to all clients after several rounds of client-server iteration. Extensive experiments using four baselines across three datasets demonstrate that our method performs favourably against state-of-the-art methods.

In this article, an X-Band (10-12 GHz) embedded multi-band LC notch filter low-noise amplifier (LNA) employing a three-stage current reuse (CR) technique and a T-type attenuator configuration is presented. The proposed notch filter is based on a LC resonant network, which has two modes of connection to the main RF path, achieving significant out-of-band suppression level of 25 dB and 50 dB in the ranges of 14-18 GHz and 26-33 GHz, respectively. The LNA is simulated using a 0.15-um GaAs pHEMT process, the simulated results show a gain exceeding 21 dB, a noise figure (NF) below 1.15 dB at 12 GHz, input and output return losses (S11and S22) are below -15 dB and below -20 dB, respectively. Additionally, it is unconditional stability in the whole frequency range. The consumption of the LNA is 64.4 mW under the 3.3-V supply voltage. The overall chip area is 1.7 mm2including all pads.

Metro passenger flow prediction is crucial for effective urban transportation management. However, its practical adoption is hindered by data silos from distributed automatic fare collection (AFC) systems, compromising prediction accuracy. While federated graph learning facilitates privacy-preserving collaboration, existing methods struggle with the unique challenges of cross-line metro passenger flow prediction, particularly in handling time-evolving spatial correlations and heterogeneous temporal correlations. To address these challenges, we present FedMetro, a novel metro passenger flow prediction system based on federated graph learning. We introduce a federated dynamic graph learning approach with cross-attention mechanisms to capture spatial-temporal correlations in passenger flow. Additionally, we propose a dynamic mask-based communication compression method to mitigate communication bottlenecks in federated inference. Extensive evaluations on three real-world metro AFC datasets demonstrate that FedMetro significantly outperforms baseline methods, achieving up to 17.08% higher accuracy while reducing federated inference communication overhead by 77.99%. Practical deployments further confirm its effectiveness in delivering accurate station-level predictions across metro lines. Our code is available at https://github.com/AlexMufeng/FedMetro.

Multi-view retrieval faces significant pressure due to the rapidly increasing multi-view information on the internet. The multi-view hashing method turns continuous features into compact information of fixed length and considerably improves retrieval efficiency. However, existing multi-view hashing methods neglect the bias produced during multi-view alignment and multi-label guidance processes. To address these issues, we introduce a novel multi-view hash method that learns compact hash codes. It first employs a multi-view double alignment module to align features from different views. Then, it utilizes a self-adjusted cross-attention fusion module to fuse these features. Finally, we propose a weighted contrastive learning module to learn more discriminative representations, smoothing the differences among all samples. Extensive experiments show that our method yields compact hash codes and outperforms state-of-the-art methods.

The goal of style transfer is to render images with the attribute dependencies of style images while maintaining the original content structure.Some recent works mainly extract the statistical information of feature maps to match the target style, but the key challenge is that the captured feature representations are too homogeneous, and cross-domain mutual exclusion may occur in the alignment process, resulting in information loss, mismatch, planarization, and so on.To this end, we propose a semi-parametric framework based on multi-perspective feature fusion and guided alignment (SPMPFA), and a backtracking loss function for content maintenance.The SPMPFA and the backgracking loss work together to capture rich presentation information while maintaining structure, thus achieving style consistency between similar fine-grained semantics and global style hierarchy.Specifically, we first use adaptive aggregation and mapping Transformer (AMTransformer) to build a cross-domain graph carrying location information inside the module and use information based on weight aggregation as associated features to guide the style alignment trend.Then, we use the feature fusion strategy to adaptively fuse the heterogeneous representation information.Finally, the content structure is maintained to the maximum extent by using backtracking loss.Qualitative and quantitative experiments demonstrate the effectiveness of our work compared to other style transfer tasks.

Flight planning, a pivotal challenge in the airline industry, strives to achieve economic and flexible scheduling of airplanes to serve designated flight itineraries. As the demand for air transportation soars, traditional planning methods can be inefficient in managing large-scale flights. Thus, we introduce Swift, a data-driven system tailored to enhance the scalability and effectiveness of flight planning. Swift primarily employs the bipartite graph model to derive optimal and economic flight plans for airlines. Our method not only minimizes the number of required planes but also ensures a balanced workload across these planes. Furthermore, Swift offers the capability of dynamic updates to flight plans in response to unexpected incidents at airports, such as bad weather conditions. Besides, Swift incorporates other functionalities like predicting future flight demand and monitoring real-time flight trajectories. Conference participants can interact with this system and explore our flight planning solution in real-world scenarios.

Shared mobility leverages under-utilized vehicles to offer on-demand transport services by sharing vehicles among users. It strives to match supply with demand via a series of data-intensive operations such as supply prediction and task assignment. However, its full potential is often compromised in practice as most shared mobility platforms operate in isolation, leading to sub-optimal resource utilization. In this demonstration, we advocate a federated approach to shared mobility, which enhances its effectiveness by enabling optimizations across platforms while retaining their autonomy. We develop privacy-preserving operators and incentive mechanisms dedicated to supply prediction and task assignment in shared mobility and implement generic interfaces that support diverse prediction and assignment algorithms. We showcase the shared mobility system with real-world ride-hailing applications.

One big challenge to federated learning is the non-IID data distribution caused by imbalanced classes. Existing federated learning approaches tend to bias towards classes containing a larger number of samples during local updates, which causes unwanted drift in the local classifiers. To address this issue, we propose a classifier debiased federated learning framework named FedCD for non-IID data. We introduce a novel hierarchical prototype contrastive learning strategy to learn fine-grained prototypes for each class. The prototypes characterize the sample distribution within each class, which helps align the features learned in the representation layer of every client's local model. At the representation layer, we use fine-grained prototypes to rebalance the class distribution on each client and rectify the classification layer of each local model. To alleviate the bias of the classification layer of the local models, we incorporate a global information distillation method to enable the local classifier to learn decoupled global classification information. We also adaptively aggregate the class-level classifiers based on their quality to reduce the impact of unreliable classes in each aggregated classifier. This mitigates the impact of client-side classifier bias on the global classifier. Comprehensive experiments conducted on various datasets show that our method, FedCD, effectively corrects classifier bias and outperforms state-of-the-art federated learning methods.

Crude oil entering the sea will be caused change in tension, viscosity, volume and etc. This process is called emulsification of crude oil. This paper found that the density of crude oil with asphalt content7% increase with the increase of emulsification and decreases gradually with temperature increment during emulsification through experiments in laboratory. Tension and viscosity of crude oil increase with emulsification, but it decrease with temperature rise. The study found that the wind speed > 3.5 m/s is conducive to crude oil volatilization, and also found that the continuous volatilization of thick oil film is greater than that of thin oil film under the same wind speed in laboratory. Under natural conditions, it was found that when the wind speeds was less than 2 m/s, the temperature had greater effect on evaporation (air temperature >32 °C ) and the volatilization increased with the increase of air temperature for crude oil. Emulsification and weathering characteristics of crude oil with different content of asphalt content are studied; it is helpful to monitor offshore oil leakage by remote sensing.

Oil spill detection algorithms for the low spectral resolution of high spatial resolution land satellite images are currently limited. Some algorithms are based on the difference of the reflectance between the seawater and the oil spill; however, a proper threshold to detect the oil spill pixel from seawater pixel is much more difficult to define. This paper proposes a new dynamic threshold oil spill detection algorithm for the Landsat ETM+ for demonstration purposes. The new MOD09A1 Version 6 surface reflectance ocean products all year round were selected for constructing the monthly priori surface reflectance database for the dynamic threshold calculations. The 6S model simulates the relationship between the surface reflectance and the apparent reflectance with different geometric parameters, atmospheric models, and aerosol optical depths. The dynamic threshold models are established based on the simulated relationship by the 6S model. The Landsat ETM+ images with oil spill that happened in the Gulf of Mexico were collected and analyzed. The results indicate that the proposed algorithm showed a better result of oil spill detection in high accuracy.