Search

We investigate the problem of sybil (fake account) detection in social networks from a graph algorithms perspective, where graph structural information is used to classify users as sybil and benign. We introduce the novel notion of user resistance to attack requests (friendship requests from sybil accounts). Building on this notion, we propose a synthetic graph data generation framework that supports various attack strategies. We then study the optimization problem where we are allowed to reveal the resistance of a subset of users with the aim to maximize the number of users which are discovered to be benign and the number of potential attack edges (connections from a sybil to a benign user). Furthermore, we devise efficient algorithms for this problem and investigate their theoretical guarantees. Finally, through a large set of experiments, we demonstrate that our proposed algorithms improve detection performance notably when applied as a preprocessing step for different sybil detection algorithms. The code and data used in this work are publicly available on GitHub https://github.com/aSafarpoor/AAMAS2025-Paper/tree/main, Comment: This is the full version of the paper accepted in The 24th International Conference on Autonomous Agents and Multiagent Systems (AAMAS 2025) with appendices

Recently, there has been an increasing interest in the Finslerian interpretation of null geodesics in the exterior regions of stationary black holes, particularly through the Zermelo navigation problem and the Randers metric. In this work, we show that recent mathematical advancements in the critical and strong Zermelo navigation problems and their connections to Kropina and Lorentz-Finsler metrics enable the extension of this Finslerian framework to encompass horizons and their interior regions. The Finslerian indicatrix, a key element of this framework, serves as an effective tool for identifying frame-dragging effects and the location of horizons and ergosurfaces. We illustrate our results with explicit physical examples, focusing on spherically symmetric and Kerr black holes. Our findings provide new insights into the "river model" of black holes, offering enhanced visual representations of null geodesics on the ergosurfaces, horizons, and within their interior regions., Comment: 9 pages, 2 figures

The increasing frequency and intensity of wildfires underscore the need for accurate predictive models to enhance wildfire management. Traditional models, such as Rothermel and FARSITE, provide foundational insights but often oversimplify the complex dynamics of wildfire spread. Advanced methods, employing sophisticated mathematical techniques, offer more precise modeling by accounting for real-world complexities and dynamic environmental factors. This paper focuses on wildfire propagation over inclined terrains and combines the Rothermel model, Huygens' principle, and advanced mathematical techniques to provide a more precise model of propagation. Environmental parameters and vegetation factors are directly incorporated into formulas and equations to improve the reliability and effectiveness of wildfire management strategies. The practical application of these results is demonstrated through MATLAB simulations, specifically examining wildfire spread under wind conditions that do not impede upwind fire advancement. The findings of this work contribute to both wildfire research and the development of more effective management strategies.

Mobile Edge Computing (MEC) assisted by Unmanned Aerial Vehicle (UAV) has been widely investigated as a promising system for future Internet-of-Things (IoT) networks. In this context, delay-sensitive tasks of IoT devices may either be processed locally or offloaded for further processing to a UAV or to the cloud. This paper, by attributing task queues to each IoT device, the UAV, and the cloud, proposes a real-time resource allocation framework in a UAV-aided MEC system. Specifically, aimed at characterizing a long-term trade-off between the time-averaged aggregate processed data (PD) and the time-averaged aggregate communication delay (CD), a resource allocation optimization problem is formulated. This problem optimizes communication and computation resources as well as the UAV motion trajectory, while guaranteeing queue stability. To address this long-term time-averaged problem, a Lyapunov optimization framework is initially leveraged to obtain an equivalent short-term optimization problem. Subsequently, we reformulate the short-term problem in a Markov Decision Process (MDP) form, where a Deep Q Network (DQN) model is trained to optimize its variables. Extensive simulations demonstrate that the proposed resource allocation scheme improves the system performance by up to 36\% compared to baseline models.

Accurate attribute extraction is critical for beauty product recommendations and building trust with customers. This remains an open problem, as existing solutions are often unreliable and incomplete. We present a system to extract beauty-specific attributes using end-to-end supervised learning based on beauty product ingredients. A key insight to our system is a novel energy-based implicit model architecture. We show that this implicit model architecture offers significant benefits in terms of accuracy, explainability, robustness, and flexibility. Furthermore, our implicit model can be easily fine-tuned to incorporate additional attributes as they become available, making it more useful in real-world applications. We validate our model on a major e-commerce skincare product catalog dataset and demonstrate its effectiveness. Finally, we showcase how ingredient-based attribute extraction contributes to enhancing the explainability of beauty recommendations., Comment: 18th ACM Conference on Recommender Systems, Workshop on Strategic and Utility-aware REcommendation

This paper focuses on designing a privacy-preserving Machine Learning (ML) inference protocol for a hierarchical setup, where clients own/generate data, model owners (cloud servers) have a pre-trained ML model, and edge servers perform ML inference on clients' data using the cloud server's ML model. Our goal is to speed up ML inference while providing privacy to both data and the ML model. Our approach (i) uses model-distributed inference (model parallelization) at the edge servers and (ii) reduces the amount of communication to/from the cloud server. Our privacy-preserving hierarchical model-distributed inference, privateMDI design uses additive secret sharing and linearly homomorphic encryption to handle linear calculations in the ML inference, and garbled circuit and a novel three-party oblivious transfer are used to handle non-linear functions. privateMDI consists of offline and online phases. We designed these phases in a way that most of the data exchange is done in the offline phase while the communication overhead of the online phase is reduced. In particular, there is no communication to/from the cloud server in the online phase, and the amount of communication between the client and edge servers is minimized. The experimental results demonstrate that privateMDI significantly reduces the ML inference time as compared to the baselines.

Unmanned aerial vehicles (UAVs) with multiple antennas have recently been explored to improve capacity in wireless networks. However, the strict energy constraint of UAVs, given their simultaneous flying and communication tasks, renders the exploration of energy-efficient multi-antenna techniques indispensable for UAVs. Meanwhile, lens antenna subarray (LAS) emerges as a promising energy-efficient solution that has not been previously harnessed for this purpose. In this paper, we propose a LAS-aided multi-antenna UAV to serve ground users in the downlink transmission of the terahertz (THz) band, utilizing rate splitting multiple access (RSMA) for effective beam division multiplexing. We formulate an optimization problem of maximizing the total system spectral efficiency (SE). This involves optimizing the UAV's transmit beamforming and the common rate of RSMA. By recasting the optimization problem into a Markov decision process (MDP), we propose a deep deterministic policy gradient (DDPG)-based resource allocation mechanism tailored to capture problem dynamics and optimize its variables. Moreover, given the UAV's frequent mobility and consequential system reconfigurations, we fortify the trained DDPG model with a meta-learning strategy, enhancing its adaptability to system variations. Numerically, more than 20\% energy efficiency gain is achieved by our proposed LAS-aided multi-antenna UAV equipped with 4 lenses, compared to a single-lens UAV. Simulations also demonstrate that at a signal-to-noise (SNR) of 10 dB, the incorporation of RSMA results in a 22\% SE enhancement over conventional orthogonal beam division multiple access. Furthermore, the overall system SE improves by 27\%, when meta-learning is employed for fine-tuning the conventional DDPG method in literature.

This paper centers around a multiple-input-multiple-output (MIMO) visible light communication (VLC) system, where an unmanned aerial vehicle (UAV) benefits from a light emitting diode (LED) array to serve photo-diode (PD)-equipped users for illumination and communication simultaneously. Concerning the battery limitation of the UAV and considerable energy consumption of the LED array, a hybrid dimming control scheme is devised at the UAV that effectively controls the number of glared LEDs and thereby mitigates the overall energy consumption. To assess the performance of this system, a radio resource allocation problem is accordingly formulated for jointly optimizing the motion trajectory, transmit beamforming and LED selection at the UAV, assuming that channel state information (CSI) is partially available. By reformulating the optimization problem in Markov decision process (MDP) form, we propose a soft actor-critic (SAC) mechanism that captures the dynamics of the problem and optimizes its parameters. Additionally, regarding the frequent mobility of the UAV and thus remarkable rearrangement of the system, we enhance the trained SAC model by integrating a meta-learning strategy that enables more adaptation to system variations. According to simulations, upgrading a single-LED UAV by an array of 10 LEDs, exhibits 47% and 34% improvements in data rate and energy efficiency, albeit at the expense of 8% more power consumption.

The multitude of makeup products available can make it challenging to find the ideal match for desired attributes. An intelligent approach for product discovery is required to enhance the makeup shopping experience to make it more convenient and satisfying. However, enabling accurate and efficient product discovery requires extracting detailed attributes like color and finish type. Our work introduces an automated pipeline that utilizes multiple customized machine learning models to extract essential material attributes from makeup product images. Our pipeline is versatile and capable of handling various makeup products. To showcase the efficacy of our pipeline, we conduct extensive experiments on eyeshadow products (both single and multi-shade ones), a challenging makeup product known for its diverse range of shapes, colors, and finish types. Furthermore, we demonstrate the applicability of our approach by successfully extending it to other makeup categories like lipstick and foundation, showcasing its adaptability and effectiveness across different beauty products. Additionally, we conduct ablation experiments to demonstrate the superiority of our machine learning pipeline over human labeling methods in terms of reliability. Our proposed method showcases its effectiveness in cross-category product discovery, specifically in recommending makeup products that perfectly match a specified outfit. Lastly, we also demonstrate the application of these material attributes in enabling virtual-try-on experiences which makes makeup shopping experience significantly more engaging., Comment: Presented in Fifth Workshop on Recommender Systems in Fashion(fashionxrecsys) of ACM Conference on Recommender Systems