Pengyuan Zhou

Pranvera Kortoçi, Yilei Liang, Pengyuan Zhou et al.

Mobile devices and the immense amount and variety of data they generate are key enablers of machine learning (ML)-based applications. Traditional ML techniques have shifted toward new paradigms such as federated (FL) and split learning (SL) to improve the protection of user's data privacy. However, these paradigms often rely on server(s) located in the edge or cloud to train computationally-heavy parts of a ML model to avoid draining the limited resource on client devices, resulting in exposing device data to such third parties. This work proposes an alternative approach to train computationally-heavy ML models in user's devices themselves, where corresponding device data resides. Specifically, we focus on GANs (generative adversarial networks) and leverage their inherent privacy-preserving attribute. We train the discriminative part of a GAN with raw data on user's devices, whereas the generative model is trained remotely (e.g., server) for which there is no need to access sensor true data. Moreover, our approach ensures that the computational load of training the discriminative model is shared among user's devices-proportional to their computation capabilities-by means of SL. We implement our proposed collaborative training scheme of a computationally-heavy GAN model in real resource-constrained devices. The results show that our system preserves data privacy, keeps a short training time, and yields same accuracy of model training in unconstrained devices (e.g., cloud). Our code can be found on https://github.com/YukariSonz/FSL-GAN

Linxin Song, Jieyu Zhang, Lechao Cheng et al. · uw

Recent developments in large language models (LLMs) have shown promise in enhancing the capabilities of natural language processing (NLP). Despite these successes, there remains a dearth of research dedicated to the NLP problem-solving abilities of LLMs. To fill the gap in this area, we present a unique benchmarking dataset, NLPBench, comprising 378 college-level NLP questions spanning various NLP topics sourced from Yale University's prior final exams. NLPBench includes questions with context, in which multiple sub-questions share the same public information, and diverse question types, including multiple choice, short answer, and math. Our evaluation, centered on LLMs such as GPT-3.5/4, PaLM-2, and LLAMA-2, incorporates advanced prompting strategies like the chain-of-thought (CoT) and tree-of-thought (ToT). Our study reveals that the effectiveness of the advanced prompting strategies can be inconsistent, occasionally damaging LLM performance, especially in smaller models like the LLAMA-2 (13b). Furthermore, our manual assessment illuminated specific shortcomings in LLMs' scientific problem-solving skills, with weaknesses in logical decomposition and reasoning notably affecting results.

Yingguang Yang, Renyu Yang, Hao Peng et al.

Social bot detection is of paramount importance to the resilience and security of online social platforms. The state-of-the-art detection models are siloed and have largely overlooked a variety of data characteristics from multiple cross-lingual platforms. Meanwhile, the heterogeneity of data distribution and model architecture makes it intricate to devise an efficient cross-platform and cross-model detection framework. In this paper, we propose FedACK, a new federated adversarial contrastive knowledge distillation framework for social bot detection. We devise a GAN-based federated knowledge distillation mechanism for efficiently transferring knowledge of data distribution among clients. In particular, a global generator is used to extract the knowledge of global data distribution and distill it into each client's local model. We leverage local discriminator to enable customized model design and use local generator for data enhancement with hard-to-decide samples. Local training is conducted as multi-stage adversarial and contrastive learning to enable consistent feature spaces among clients and to constrain the optimization direction of local models, reducing the divergences between local and global models. Experiments demonstrate that FedACK outperforms the state-of-the-art approaches in terms of accuracy, communication efficiency, and feature space consistency.

Xu Zheng, Yunhao Luo, Pengyuan Zhou et al.

In this paper, we tackle a new problem: how to transfer knowledge from the pre-trained cumbersome yet well-performed CNN-based model to learn a compact Vision Transformer (ViT)-based model while maintaining its learning capacity? Due to the completely different characteristics of ViT and CNN and the long-existing capacity gap between teacher and student models in Knowledge Distillation (KD), directly transferring the cross-model knowledge is non-trivial. To this end, we subtly leverage the visual and linguistic-compatible feature character of ViT (i.e., student), and its capacity gap with the CNN (i.e., teacher) and propose a novel CNN-to-ViT KD framework, dubbed C2VKD. Importantly, as the teacher's features are heterogeneous to those of the student, we first propose a novel visual-linguistic feature distillation (VLFD) module that explores efficient KD among the aligned visual and linguistic-compatible representations. Moreover, due to the large capacity gap between the teacher and student and the inevitable prediction errors of the teacher, we then propose a pixel-wise decoupled distillation (PDD) module to supervise the student under the combination of labels and teacher's predictions from the decoupled target and non-target classes. Experiments on three semantic segmentation benchmark datasets consistently show that the increment of mIoU of our method is over 200% of the SoTA KD methods

Anish K. Vallapuram, Pengyuan Zhou, Young D. Kwon et al.

Federated learning alleviates the privacy risk in distributed learning by transmitting only the local model updates to the central server. However, it faces challenges including statistical heterogeneity of clients' datasets and resource constraints of client devices, which severely impact the training performance and user experience. Prior works have tackled these challenges by combining personalization with model compression schemes including quantization and pruning. However, the pruning is data-dependent and thus must be done on the client side which requires considerable computation cost. Moreover, the pruning normally trains a binary supermask $\in \{0, 1\}$ which significantly limits the model capacity yet with no computation benefit. Consequently, the training requires high computation cost and a long time to converge while the model performance does not pay off. In this work, we propose HideNseek which employs one-shot data-agnostic pruning at initialization to get a subnetwork based on weights' synaptic saliency. Each client then optimizes a sign supermask $\in \{-1, +1\}$ multiplied by the unpruned weights to allow faster convergence with the same compression rates as state-of-the-art. Empirical results from three datasets demonstrate that compared to state-of-the-art, HideNseek improves inferences accuracies by up to 40.6\% while reducing the communication cost and training time by up to 39.7\% and 46.8\% respectively.

Pengyuan Zhou

Incorporating artificial intelligence (AI) technology, particularly large language models (LLMs), is becoming increasingly vital for developing immersive and interactive metaverse experiences. GPT, a representative LLM developed by OpenAI, is leading LLM development and gaining attention for its potential in building the metaverse. The article delves into the pros and cons of utilizing GPT for metaverse-based education, entertainment, personalization, and support. Dynamic and personalized experiences are possible with this technology, but there are also legitimate privacy, bias, and ethical issues to consider. This article aims to help readers understand the possible influence of GPT, according to its unique technological advantages, on the metaverse and how it may be used to effectively create a more immersive and engaging virtual environment by evaluating these opportunities and obstacles.

Yao Lu, Pengyuan Zhou, Yong Liao et al.

Urban anomaly predictions, such as traffic accident prediction and crime prediction, are of vital importance to smart city security and maintenance. Existing methods typically use deep learning to capture the intra-dependencies in spatial and temporal dimensions. However, numerous key challenges remain unsolved, for instance, sparse zero-inflated data due to urban anomalies occurring with low frequency (which can lead to poor performance on real-world datasets), and both intra- and inter-dependencies of abnormal patterns across spatial, temporal, and semantic dimensions. Moreover, a unified approach to predict multiple kinds of anomaly is left to explore. In this paper, we propose STS to jointly capture the intra- and inter-dependencies between the patterns and the influential factors in three dimensions. Further, we use a multi-task prediction module with a customized loss function to solve the zero-inflated issue. To verify the effectiveness of the model, we apply it to two urban anomaly prediction tasks, crime prediction and traffic accident risk prediction, respectively. Experiments on two application scenarios with four real-world datasets demonstrate the superiority of STS, which outperforms state-of-the-art methods in the mean absolute error and the root mean square error by 37.88% and 18.10% on zero-inflated datasets, and, 60.32% and 37.28% on non-zero datasets, respectively.

Jie Li, Zhixin Li, Zhi Liu et al.

Volumetric video, also known as hologram video, is a novel medium that portrays natural content in Virtual Reality (VR), Augmented Reality (AR), and Mixed Reality (MR). It is expected to be the next-gen video technology and a prevalent use case for 5G and beyond wireless communication. Considering that each user typically only watches a section of the volumetric video, known as the viewport, it is essential to have precise viewport prediction for optimal performance. However, research on this topic is still in its infancy. In the end, this paper presents and proposes a novel approach, named Saliency and Trajectory Viewport Prediction (STVP), which aims to improve the precision of viewport prediction in volumetric video streaming. The STVP extensively utilizes video saliency information and viewport trajectory. To our knowledge, this is the first comprehensive study of viewport prediction in volumetric video streaming. In particular, we introduce a novel sampling method, Uniform Random Sampling (URS), to reduce computational complexity while still preserving video features in an efficient manner. Then we present a saliency detection technique that incorporates both spatial and temporal information for detecting static, dynamic geometric, and color salient regions. Finally, we intelligently fuse saliency and trajectory information to achieve more accurate viewport prediction. We conduct extensive simulations to evaluate the effectiveness of our proposed viewport prediction methods using state-of-the-art volumetric video sequences. The experimental results show the superiority of the proposed method over existing schemes. The dataset and source code will be publicly accessible after acceptance.

Hengwei Xu, Yong Liao, Haiyong Xie et al.

The rapidly enlarging neural network models are becoming increasingly challenging to run on a single device. Hence model parallelism over multiple devices is critical to guarantee the efficiency of training large models. Recent proposals fall short either in long processing time or poor performance. Therefore, we propose Celeritas, a fast framework for optimizing device placement for large models. Celeritas employs a simple but efficient model parallelization strategy in the Standard Evaluation, and generates placement policies through a series of scheduling algorithms. We conduct experiments to deploy and evaluate Celeritas on numerous large models. The results show that Celeritas not only reduces the placement policy generation time by 26.4\% but also improves the model running time by 34.2\% compared to most advanced methods.

Yan Shi, Yao Tian, Chengwei Tong et al.

Social network plays an important role in propagating people's viewpoints, emotions, thoughts, and fears. Notably, following lockdown periods during the COVID-19 pandemic, the issue of depression has garnered increasing attention, with a significant portion of individuals resorting to social networks as an outlet for expressing emotions. Using deep learning techniques to discern potential signs of depression from social network messages facilitates the early identification of mental health conditions. Current efforts in detecting depression through social networks typically rely solely on analyzing the textual content, overlooking other potential information. In this work, we conduct a thorough investigation that unveils a strong correlation between depression and negative emotional states. The integration of such associations as external knowledge can provide valuable insights for detecting depression. Accordingly, we propose a multi-task training framework, DeSK, which utilizes shared sentiment knowledge to enhance the efficacy of depression detection. Experiments conducted on both Chinese and English datasets demonstrate the cross-lingual effectiveness of DeSK.

Yihang Lin, Pengyuan Zhou, Zhiqian Wu et al.

Federated learning allows clients to collaboratively train a global model without uploading raw data for privacy preservation. This feature, i.e., the inability to review participants' datasets, has recently been found responsible for federated learning's vulnerability in the face of backdoor attacks. Existing defense methods fall short from two perspectives: 1) they consider only very specific and limited attacker models and unable to cope with advanced backdoor attacks, such as distributed backdoor attacks, which break down the global trigger into multiple distributed triggers. 2) they conduct detection based on model granularity thus the performance gets impacted by the model dimension. To address these challenges, we propose Federated Layer Detection (FLD), a novel model filtering approach for effectively defending against backdoor attacks. FLD examines the models based on layer granularity to capture the complete model details and effectively detect potential backdoor models regardless of model dimension. We provide theoretical analysis and proof for the convergence of FLD. Extensive experiments demonstrate that FLD effectively mitigates state-of-the-art backdoor attacks with negligible impact on the accuracy of the primary task.

Haoran Li, XiaoLu Li, Yihang Lin et al.

Video prediction is a complex time-series forecasting task with great potential in many use cases. However, traditional methods prioritize accuracy and overlook slow prediction speeds due to complex model structures, redundant information, and excessive GPU memory consumption. These methods often predict frames sequentially, making acceleration difficult and limiting their applicability in real-time scenarios like danger prediction and warning.Therefore, we propose a transformer-based keypoint prediction neural network (TKN). TKN extracts dynamic content from video frames in an unsupervised manner, reducing redundant feature computation. And, TKN uses an acceleration matrix to reduce the computational cost of attention and employs a parallel computing structure for prediction acceleration. To the best of our knowledge, TKN is the first real-time video prediction solution that achieves a prediction rate of 1,176 fps, significantly reducing computation costs while maintaining other performance. Qualitative and quantitative experiments on multiple datasets have demonstrated the superiority of our method, suggesting that TKN has great application potential.

Haoran Li, Long Ma, Haolin Shi et al.

The current GAN inversion methods typically can only edit the appearance and shape of a single object and background while overlooking spatial information. In this work, we propose a 3D editing framework, 3D-GOI, to enable multifaceted editing of affine information (scale, translation, and rotation) on multiple objects. 3D-GOI realizes the complex editing function by inverting the abundance of attribute codes (object shape/appearance/scale/rotation/translation, background shape/appearance, and camera pose) controlled by GIRAFFE, a renowned 3D GAN. Accurately inverting all the codes is challenging, 3D-GOI solves this challenge following three main steps. First, we segment the objects and the background in a multi-object image. Second, we use a custom Neural Inversion Encoder to obtain coarse codes of each object. Finally, we use a round-robin optimization algorithm to get precise codes to reconstruct the image. To the best of our knowledge, 3D-GOI is the first framework to enable multifaceted editing on multiple objects. Both qualitative and quantitative experiments demonstrate that 3D-GOI holds immense potential for flexible, multifaceted editing in complex multi-object scenes.Our project and code are released at https://3d-goi.github.io .

Long Ma, Zhiyuan Yan, Qinglang Guo et al.

The escalating quality of video generated by advanced video generation methods results in new security challenges, while there have been few relevant research efforts: 1) There is no open-source dataset for generated video detection, 2) No generated video detection method has been proposed so far. To this end, we propose an open-source dataset and a detection method for generated video for the first time. First, we propose a scalable dataset consisting of 964 prompts, covering various forgery targets, scenes, behaviors, and actions, as well as various generation models with different architectures and generation methods, including the most popular commercial models like OpenAI's Sora and Google's Veo. Second, we found via probing experiments that spatial artifact-based detectors lack generalizability. Hence, we propose a simple yet effective \textbf{de}tection model based on \textbf{f}rame \textbf{co}nsistency (\textbf{DeCoF}), which focuses on temporal artifacts by eliminating the impact of spatial artifacts during feature learning. Extensive experiments demonstrate the efficacy of DeCoF in detecting videos generated by unseen video generation models and confirm its powerful generalizability across several commercially proprietary models.

Wei Tang, Zhiqian Wu, Yixin Cao et al.

Knowledge graph completion (KGC) aims to predict missing facts in knowledge graphs (KGs), which is crucial as modern KGs remain largely incomplete. While training KGC models on multiple aligned KGs can improve performance, previous methods that rely on transferring raw data among KGs raise privacy concerns. To address this challenge, we propose a new federated learning framework that implicitly aggregates knowledge from multiple KGs without demanding raw data exchange and entity alignment. We treat each KG as a client that trains a local language model through textbased knowledge representation learning. A central server then aggregates the model weights from clients. As natural language provides a universal representation, the same knowledge thus has similar semantic representations across KGs. As such, the aggregated language model can leverage complementary knowledge from multilingual KGs without demanding raw user data sharing. Extensive experiments on a benchmark dataset demonstrate that our method substantially improves KGC on multilingual KGs, achieving comparable performance to state-of-the-art alignment-based models without requiring any labeled alignments or raw user data sharing. Our codes will be publicly available.

Haoran Li, Haolin Shi, Wenli Zhang et al.

Text-to-3D scene generation holds immense potential for the gaming, film, and architecture sectors. Despite significant progress, existing methods struggle with maintaining high quality, consistency, and editing flexibility. In this paper, we propose DreamScene, a 3D Gaussian-based novel text-to-3D scene generation framework, to tackle the aforementioned three challenges mainly via two strategies. First, DreamScene employs Formation Pattern Sampling (FPS), a multi-timestep sampling strategy guided by the formation patterns of 3D objects, to form fast, semantically rich, and high-quality representations. FPS uses 3D Gaussian filtering for optimization stability, and leverages reconstruction techniques to generate plausible textures. Second, DreamScene employs a progressive three-stage camera sampling strategy, specifically designed for both indoor and outdoor settings, to effectively ensure object-environment integration and scene-wide 3D consistency. Last, DreamScene enhances scene editing flexibility by integrating objects and environments, enabling targeted adjustments. Extensive experiments validate DreamScene's superiority over current state-of-the-art techniques, heralding its wide-ranging potential for diverse applications. Code and demos will be released at https://dreamscene-project.github.io .

Pengyuan Zhou, Lin Wang, Zhi Liu et al.

This paper offers an insightful examination of how currently top-trending AI technologies, i.e., generative artificial intelligence (Generative AI) and large language models (LLMs), are reshaping the field of video technology, including video generation, understanding, and streaming. It highlights the innovative use of these technologies in producing highly realistic videos, a significant leap in bridging the gap between real-world dynamics and digital creation. The study also delves into the advanced capabilities of LLMs in video understanding, demonstrating their effectiveness in extracting meaningful information from visual content, thereby enhancing our interaction with videos. In the realm of video streaming, the paper discusses how LLMs contribute to more efficient and user-centric streaming experiences, adapting content delivery to individual viewer preferences. This comprehensive review navigates through the current achievements, ongoing challenges, and future possibilities of applying Generative AI and LLMs to video-related tasks, underscoring the immense potential these technologies hold for advancing the field of video technology related to multimedia, networking, and AI communities.

Xu Zheng, Pengyuan Zhou, Athanasios V. Vasilakos et al.

In this paper, we address the challenging source-free unsupervised domain adaptation (SFUDA) for pinhole-to-panoramic semantic segmentation, given only a pinhole image pre-trained model (i.e., source) and unlabeled panoramic images (i.e., target). Tackling this problem is non-trivial due to three critical challenges: 1) semantic mismatches from the distinct Field-of-View (FoV) between domains, 2) style discrepancies inherent in the UDA problem, and 3) inevitable distortion of the panoramic images. To tackle these problems, we propose 360SFUDA++ that effectively extracts knowledge from the source pinhole model with only unlabeled panoramic images and transfers the reliable knowledge to the target panoramic domain. Specifically, we first utilize Tangent Projection (TP) as it has less distortion and meanwhile slits the equirectangular projection (ERP) to patches with fixed FoV projection (FFP) to mimic the pinhole images. Both projections are shown effective in extracting knowledge from the source model. However, as the distinct projections make it less possible to directly transfer knowledge between domains, we then propose Reliable Panoramic Prototype Adaptation Module (RP2AM) to transfer knowledge at both prediction and prototype levels. RP$^2$AM selects the confident knowledge and integrates panoramic prototypes for reliable knowledge adaptation. Moreover, we introduce Cross-projection Dual Attention Module (CDAM), which better aligns the spatial and channel characteristics across projections at the feature level between domains. Both knowledge extraction and transfer processes are synchronously updated to reach the best performance. Extensive experiments on the synthetic and real-world benchmarks, including outdoor and indoor scenarios, demonstrate that our 360SFUDA++ achieves significantly better performance than prior SFUDA methods.

Kun Lan, Haoran Li, Haolin Shi et al.

Recently, 3D Gaussian, as an explicit 3D representation method, has demonstrated strong competitiveness over NeRF (Neural Radiance Fields) in terms of expressing complex scenes and training duration. These advantages signal a wide range of applications for 3D Gaussians in 3D understanding and editing. Meanwhile, the segmentation of 3D Gaussians is still in its infancy. The existing segmentation methods are not only cumbersome but also incapable of segmenting multiple objects simultaneously in a short amount of time. In response, this paper introduces a 3D Gaussian segmentation method implemented with 2D segmentation as supervision. This approach uses input 2D segmentation maps to guide the learning of the added 3D Gaussian semantic information, while nearest neighbor clustering and statistical filtering refine the segmentation results. Experiments show that our concise method can achieve comparable performances on mIOU and mAcc for multi-object segmentation as previous single-object segmentation methods.

Buyun He, Yingguang Yang, Qi Wu et al.

Detecting social bots has evolved into a pivotal yet intricate task, aimed at combating the dissemination of misinformation and preserving the authenticity of online interactions. While earlier graph-based approaches, which leverage topological structure of social networks, yielded notable outcomes, they overlooked the inherent dynamicity of social networks -- In reality, they largely depicted the social network as a static graph and solely relied on its most recent state. Due to the absence of dynamicity modeling, such approaches are vulnerable to evasion, particularly when advanced social bots interact with other users to camouflage identities and escape detection. To tackle these challenges, we propose BotDGT, a novel framework that not only considers the topological structure, but also effectively incorporates dynamic nature of social network. Specifically, we characterize a social network as a dynamic graph. A structural module is employed to acquire topological information from each historical snapshot. Additionally, a temporal module is proposed to integrate historical context and model the evolving behavior patterns exhibited by social bots and legitimate users. Experimental results demonstrate the superiority of BotDGT against the leading methods that neglected the dynamic nature of social networks in terms of accuracy, recall, and F1-score.

Qinglong Huang, Haoran Li, Yong Liao et al.

Neural Radiance Field (NeRF) has been proposed as an innovative advancement in 3D reconstruction techniques. However, little research has been conducted on the issues of information confidentiality and security to NeRF, such as steganography. Existing NeRF steganography solutions have shortcomings in low steganography quality, model weight damage, and limited amount of steganographic information. This paper proposes Noise-NeRF, a novel NeRF steganography method employing Adaptive Pixel Selection strategy and Pixel Perturbation strategy to improve the quality and efficiency of steganography via trainable noise. Extensive experiments validate the state-of-the-art performances of Noise-NeRF on both steganography quality and rendering quality, as well as effectiveness in super-resolution image steganography.

Wei Tang, Yixin Cao, Jiahao Ying et al.

Retrieval-Augmented Generation (RAG) is an effective solution to supplement necessary knowledge to large language models (LLMs). Targeting its bottleneck of retriever performance, "generate-then-read" pipeline is proposed to replace the retrieval stage with generation from the LLM itself. Although promising, this research direction is underexplored and still cannot work in the scenario when source knowledge is given. In this paper, we formalize a general "A + B" framework with varying combinations of foundation models and types for systematic investigation. We explore the efficacy of the base and chat versions of LLMs and found their different functionalities suitable for generator A and reader B, respectively. Their combinations consistently outperform single models, especially in complex scenarios. Furthermore, we extend the application of the "A + B" framework to scenarios involving source documents through continuous learning, enabling the direct integration of external knowledge into LLMs. This approach not only facilitates effective acquisition of new knowledge but also addresses the challenges of safety and helpfulness post-adaptation. The paper underscores the versatility of the "A + B" framework, demonstrating its potential to enhance the practical application of LLMs across various domains.

Xu Zheng, Pengyuan Zhou, Athanasios V. Vasilakos et al.

This paper addresses an interesting yet challenging problem -- source-free unsupervised domain adaptation (SFUDA) for pinhole-to-panoramic semantic segmentation -- given only a pinhole image-trained model (i.e., source) and unlabeled panoramic images (i.e., target). Tackling this problem is nontrivial due to the semantic mismatches, style discrepancies, and inevitable distortion of panoramic images. To this end, we propose a novel method that utilizes Tangent Projection (TP) as it has less distortion and meanwhile slits the equirectangular projection (ERP) with a fixed FoV to mimic the pinhole images. Both projections are shown effective in extracting knowledge from the source model. However, the distinct projection discrepancies between source and target domains impede the direct knowledge transfer; thus, we propose a panoramic prototype adaptation module (PPAM) to integrate panoramic prototypes from the extracted knowledge for adaptation. We then impose the loss constraints on both predictions and prototypes and propose a cross-dual attention module (CDAM) at the feature level to better align the spatial and channel characteristics across the domains and projections. Both knowledge extraction and transfer processes are synchronously updated to reach the best performance. Extensive experiments on the synthetic and real-world benchmarks, including outdoor and indoor scenarios, demonstrate that our method achieves significantly better performance than prior SFUDA methods for pinhole-to-panoramic adaptation.

Hao Ai, Zidong Cao, Haonan Lu et al.

360 images, with a field-of-view (FoV) of 180x360, provide immersive and realistic environments for emerging virtual reality (VR) applications, such as virtual tourism, where users desire to create diverse panoramic scenes from a narrow FoV photo they take from a viewpoint via portable devices. It thus brings us to a technical challenge: `How to allow the users to freely create diverse and immersive virtual scenes from a narrow FoV image with a specified viewport?' To this end, we propose a transformer-based 360 image outpainting framework called Dream360, which can generate diverse, high-fidelity, and high-resolution panoramas from user-selected viewports, considering the spherical properties of 360 images. Compared with existing methods, e.g., [3], which primarily focus on inputs with rectangular masks and central locations while overlooking the spherical property of 360 images, our Dream360 offers higher outpainting flexibility and fidelity based on the spherical representation. Dream360 comprises two key learning stages: (I) codebook-based panorama outpainting via Spherical-VQGAN (S-VQGAN), and (II) frequency-aware refinement with a novel frequency-aware consistency loss. Specifically, S-VQGAN learns a sphere-specific codebook from spherical harmonic (SH) values, providing a better representation of spherical data distribution for scene modeling. The frequency-aware refinement matches the resolution and further improves the semantic consistency and visual fidelity of the generated results. Our Dream360 achieves significantly lower Frechet Inception Distance (FID) scores and better visual fidelity than existing methods. We also conducted a user study involving 15 participants to interactively evaluate the quality of the generated results in VR, demonstrating the flexibility and superiority of our Dream360 framework.

Xiaorui Jiang, Yu Gao, Hengwei Xu et al.

Federated Learning (FL) allows several clients to cooperatively train machine learning models without disclosing the raw data. In practical applications, asynchronous FL (AFL) can address the straggler effect compared to synchronous FL. However, Non-IID data and stale models pose significant challenges to AFL, as they can diminish the practicality of the global model and even lead to training failures. In this work, we propose a novel AFL framework called Federated Historical Learning (FedHist), which effectively addresses the challenges posed by both Non-IID data and gradient staleness based on the concept of knowledge rumination. FedHist enhances the stability of local gradients by performing weighted fusion with historical global gradients cached on the server. Relying on hindsight, it assigns aggregation weights to each participant in a multi-dimensional manner during each communication round. To further enhance the efficiency and stability of the training process, we introduce an intelligent $\ell_2$-norm amplification scheme, which dynamically regulates the learning progress based on the $\ell_2$-norms of the submitted gradients. Extensive experiments indicate FedHist outperforms state-of-the-art methods in terms of convergence performance and test accuracy.

Pengyuan Zhou, Benjamin Finley, Lik-Hang Lee et al.

AI plays a key role in current cyberspace and future immersive ecosystems that pinpoint user experiences. Thus, the trustworthiness of such AI systems is vital as failures in these systems can cause serious user harm. Although there are related works on exploring trustworthy AI (TAI) metrics in the current cyberspace, ecosystems towards user-centered services, such as the metaverse, are much more complicated in terms of system performance and user experience assessment, thus posing challenges for the applicability of existing approaches. Thus, we give an overlook on fairness, privacy and robustness, across the historical path from existing approaches. Eventually, we propose a research agenda towards systematic yet user-centered TAI in immersive ecosystems.

Wenxiao Zhang, Sikun Lin, Farshid Hassani Bijarbooneh et al.

The computational capabilities of recent mobile devices enable the processing of natural features for Augmented Reality (AR), but the scalability is still limited by the devices' computation power and available resources. In this paper, we propose EdgeXAR, a mobile AR framework that utilizes the advantages of edge computing through task offloading to support flexible camera-based AR interaction. We propose a hybrid tracking system for mobile devices that provides lightweight tracking with 6 Degrees of Freedom and hides the offloading latency from users' perception. A practical, reliable and unreliable communication mechanism is used to achieve fast response and consistency of crucial information. We also propose a multi-object image retrieval pipeline that executes fast and accurate image recognition tasks on the cloud and edge servers. Extensive experiments are carried out to evaluate the performance of EdgeXAR by building mobile AR Apps upon it. Regarding the Quality of Experience (QoE), the mobile AR Apps powered by EdgeXAR framework run on average at the speed of 30 frames per second with precise tracking of only 1~2 pixel errors and accurate image recognition of at least 97% accuracy. As compared to Vuforia, one of the leading commercial AR frameworks, EdgeXAR transmits 87% less data while providing a stable 30 FPS performance and reducing the offloading latency by 50 to 70% depending on the transmission medium. Our work facilitates the large-scale deployment of AR as the next generation of ubiquitous interfaces.

Pengyuan Zhou, Pei Fang, Pan Hui

Federated learning has attracted attention in recent years for collaboratively training data on distributed devices with privacy-preservation. The limited network capacity of mobile and IoT devices has been seen as one of the major challenges for cross-device federated learning. Recent solutions have been focusing on threshold-based client selection schemes to guarantee the communication efficiency. However, we find this approach can cause biased client selection and results in deteriorated performance. Moreover, we find that the challenge of network limit may be overstated in some cases and the packet loss is not always harmful. In this paper, we explore the loss tolerant federated learning (LT-FL) in terms of aggregation, fairness, and personalization. We use ThrowRightAway (TRA) to accelerate the data uploading for low-bandwidth-devices by intentionally ignoring some packet losses. The results suggest that, with proper integration, TRA and other algorithms can together guarantee the personalization and fairness performance in the face of packet loss below a certain fraction (10%-30%).

Pengyuan Zhou, Pranvera Kortoci, Yui-Pan Yau et al.

Connected vehicles, whether equipped with advanced driver-assistance systems or fully autonomous, require human driver supervision and are currently constrained to visual information in their line-of-sight. A cooperative perception system among vehicles increases their situational awareness by extending their perception range. Existing solutions focus on improving perspective transformation and fast information collection. However, such solutions fail to filter out large amounts of less relevant data and thus impose significant network and computation load. Moreover, presenting all this less relevant data can overwhelm the driver and thus actually hinder them. To address such issues, we present Augmented Informative Cooperative Perception (AICP), the first fast-filtering system which optimizes the informativeness of shared data at vehicles to improve the fused presentation. To this end, an informativeness maximization problem is presented for vehicles to select a subset of data to display to their drivers. Specifically, we propose (i) a dedicated system design with custom data structure and lightweight routing protocol for convenient data encapsulation, fast interpretation and transmission, and (ii) a comprehensive problem formulation and efficient fitness-based sorting algorithm to select the most valuable data to display at the application layer. We implement a proof-of-concept prototype of AICP with a bandwidth-hungry, latency-constrained real-life augmented reality application. The prototype adds only 12.6 milliseconds of latency to a current informativeness-unaware system. Next, we test the networking performance of AICP at scale and show that ACIP effectively filters out less relevant packets and decreases the channel busy time.

Pengyuan Zhou, Xianfu Chen, Zhi Liu et al.

The Internet of Vehicles (IoV) enables real-time data exchange among vehicles and roadside units and thus provides a promising solution to alleviate traffic jams in the urban area. Meanwhile, better traffic management via efficient traffic light control can benefit the IoV as well by enabling a better communication environment and decreasing the network load. As such, IoV and efficient traffic light control can formulate a virtuous cycle. Edge computing, an emerging technology to provide low-latency computation capabilities at the edge of the network, can further improve the performance of this cycle. However, while the collected information is valuable, an efficient solution for better utilization and faster feedback has yet to be developed for edge-empowered IoV. To this end, we propose a Decentralized Reinforcement Learning at the Edge for traffic light control in the IoV (DRLE). DRLE exploits the ubiquity of the IoV to accelerate the collection of traffic data and its interpretation towards alleviating congestion and providing better traffic light control. DRLE operates within the coverage of the edge servers and uses aggregated data from neighboring edge servers to provide city-scale traffic light control. DRLE decomposes the highly complex problem of large area control. into a decentralized multi-agent problem. We prove its global optima with concrete mathematical reasoning. The proposed decentralized reinforcement learning algorithm running at each edge node adapts the traffic lights in real time. We conduct extensive evaluations and demonstrate the superiority of this approach over several state-of-the-art algorithms.