Search

A systematic method for designing reconfigurable compact multiple-input multiple-output (MIMO) antennas is described. This method is based on the analysis of a general loaded N - port structure in beamspace domain and is use-ful in optimizing various compact MIMO antennas for next-generation wireless communication. In this method, multiple streams of symbols are mapped onto a set of orthogonal far-field radiation patterns. Then by finding the required loads, the desired radiation patterns can be excited to transmit symbols. Leveraging this method, in this paper, we focus on two MIMO antenna configurations, including single-radio-frequency (RF) electronically steerable parasitic array radiator (ESPAR) and multi-RF compact antenna, and provide efficient optimization methods to find the optimal MIMO antenna design with load values. Simulation results show that by using the proposed method, the single-RF ESPAR can significantly improve the energy efficiency (EE) while achieving close spectral efficiency (SE) when compared with conventional MIMO antennas. In addition for the multi-RF compact antenna, the mutual coupling can be efficiently suppressed by generating orthogonal radiation patterns at multiple ports and the resulting SE can achieve the channel capacity bound. © 2023 IEEE.

Cardiotocography (CTG) is a clinical procedure that is used to track and gauge the severity of fetal distress. Although CTG is the most often used equipment to monitor and assess the health of the fetus, the high rate of false positive results due to visual interpretation significantly contributes to needless surgical delivery or delayed intervention. In this study, a novel approach is introduced where both printing CTG paper is digitized and a machine learning approach is employed to detect the abnormality in the digitized CTG signal. Image processing-based preprocessing steps are employed to make the printing of CTG paper more convenient to extract the CTG signal. Various signal-processing techniques are used to calibrate the extracted CTG signal. Then, Empirical Mode Decomposition (EMD) is used to decompose the CTG signal into its frequency components and instantaneous frequency and spectral entropy features are extracted. After feature normalization and feature selection with ReliefF algorithm, support vector machines (SVM) is used for the classification of the normal and abnormal classes. A novel dataset is used in the experimental works and various performance evaluation metrics are used for the evaluation of the achievement of the proposed method. 10-fold cross-validation-based experiments show that the proposed method is quite efficient in abnormality detection in printing CTG papers where an average accuracy score of around 90.0% is produced.

The robotic wearable exoskeletons have been developed due to various advantages offered by these devices. These advantages are manifested by integrating the human and a robot into one system under the user's control, which motivated researchers to develop different exoskeletons through years. However, several advances in exoskeleton; still, dealing with the various technical challenges in designing these impressive devices is inevitably. This paper aims to introduce informative resources and quick guidance of various technical challenges as such information is critical for exoskeleton development. The constructive discussion is intended to encourage researchers, innovators and academia to be aware of these challenges. Finally, the contemporary research gaps with various challenges have been highlighted, which remain to be solved as well as some future directions in this field that will have far-reaching effects on developing exoskeletons.

We demonstrate power-scaling of nonlinear frequency conversion in fibres via dispersive wave generation using higher order modes. By pumping in the LP0,m modes at 1030 nm we show wavelength selectability between 550 and 800 nm with pulse energies up to tens of nJ.

In the past two decades, there has been a consider-able increase in use of computer-based exams. Currently there is a lack of comprehensive understanding about equivalence between paper-based and computer-based exams in engineering and computing education. This study utilised a systematic literature review (using PRISMA) to synthesise the literature regarding paper and computer-based testing in engineering and computing education between 2010-2020 (inclusive). The research database that was used was Scopus. Publications were first screened by title, abstract and keywords. This was followed by screening the full text of each publication. Twenty-four papers were included in the final analysis. Outcomes demonstrated no clear result regarding equivalency between computer-based and paper-based exams. Similarly, many studies reported that students preferred computer-based exams, but this was not ubiquitous. Educators should consider these factors when considering transitioning from paper to computer-based testing.

Reconstructing simplicial signals, e.g., signals defined on nodes, edges, triangles, etc., of a network, from (partial) noisy observation is of interest in water/traffic flow estimation or currency exchange markets. Typically, this concerns solving a regularised problem w.r.t. the l2 norm of the divergence or the curl of the signal, i.e., the netflows at nodes and in triangles. Realworld simplicial signals are intrinsically divergence- or curl-free, which makes l2 regularizers inapplicable. To overcome this, we develop a simplicial trend filter (STF) by regularising the total divergence and the curl via their l1 norm. By tuning two scalars, the STF can reduce independently the divergence and curl much more than smooth filtering, leading to a better reconstructed signal. The SFT is a convex problem and can be solved by fast iterative algorithms. We apply the SFT to interpolation and denoising tasks in forex and music/artist transition recordings and show its superior performance to alternatives., Green Open Access added to TU Delft Institutional Repository 'You share, we take care!' - Taverne project https://www.openaccess.nl/en/you-share-we-take-care Otherwise as indicated in the copyright section: the publisher is the copyright holder of this work and the author uses the Dutch legislation to make this work public. Funding Information: Emails: m.yang-2, e.isufi-1@tudelft.nl. This work is supported by the TU Delft AI Labs Programme., Multimedia Computing

Theodolite is a framework for benchmarking the scalability of cloud-native applications. It automates deployment and monitoring of a cloud-native application for different load intensities and provisioned cloud resources and assesses whether specified service level objectives (SLOs) are fulfilled. Provided as a Kubernetes Operator, Theodolite allows defining, sharing, and archiving benchmarks and experiment configurations in declarative files. We demonstrate Theodolite’s benchmarking method and show how researchers and cloud engineers can execute existing scalability benchmarks or define new ones with Theodolite.

Theodolite is a framework for benchmarking the scalability of cloud-native applications. It automates deployment and monitoring of a cloud-native application for different load intensities and provisioned cloud resources and assesses whether specified service level objectives (SLOs) are fulfilled. Provided as a Kubernetes Operator, Theodolite allows defining, sharing, and archiving benchmarks and experiment configurations in declarative files. We demonstrate Theodolite’s benchmarking method and show how researchers and cloud engineers can execute existing scalability benchmarks or define new ones with Theodolite.

Traveling-wave tube amplifiers (TWTAs) are common power amplifiers in satellite communications. Saturation and warping effects caused by the nonlinearity make constellation shaping a non-trivial task, when nonlinear amplifiers are involved in the channel. Constellation shaping optimizes geometry and probability of occurrence of constellation points to maximize the mutual information. This paper newly introduces constellation shaping using neural networks to satellite communications by showing its benefits on channels with TWTAs. We study how the TWTA nonlinearity impacts learned constellations both in terms of geometry and point probability of occurrence. The peak-power constraint introduced by TWTA saturation leads to a decreased impact of probabilistic shaping compared to geometric shaping which is illustrated using mutual information curves and displayed constellations. The tuning of the input backoff (IBO) of the TWTA causes an additional trade-off between transmit power at the cost of more severe nonlinear effects. This paper makes a new contribution by training IBO and transmit constellation jointly for TWTAs. By displaying constellations at different IBOs the interdependency of constellation shaping and IBO optimization is motivated. Total degradation analysis on mutual information curves resulting from the cross-entropy loss of the autoencoder illustrates the impact of joint IBO training and constellation shaping. A study of coded bit error rate performance and comparison to conventional constellations with memoryless predistortion emphasizes the effectiveness of training IBO and constellation jointly using autoencoders.

This feasibility study deals with Air to Ground (ATG) communications between a 5G type Terrestrial Mobile Communication Network (TMCN) and an User Equipment (UE) installed onboard a passenger aircraft in approach to an airport. The network model is based on real gNodeB (gNB) deployment data and optimized for serving terrestrial UEs. The evaluation concerns a scenario in which the passenger aircraft is supported by a remote operator via an ATG link to assist its safe landing. To mitigate the interference expected to arise in ATG communications, the aircraft is assumed to be equipped with an Active Electronically Scanned Array (AESA). The simulation results discussed in this paper address the Signal to Interference power Ratio (SIR), link availability, and handover process during ATG communications along representative approach trajectories for different AESA configurations and levels of support by the TMCN. It is shown that SIR above 30 dB and link availability upto 97.7% can be achieved by using an 8 x 8 AESA at the aircraft when the network provides support in the form of dedicated radio resources. It is observed that the frequency of handovers increases with improving link quality.

With the recent development of the Virtual Reality (VR) industry, the increasing number of VR users pushes the demand for the massive production of immersive and expressive VR scenes in related industries. However, creating expressive VR scenes involves the reasonable organization of various digital content to express a coherent and logical theme, which is time-consuming and labor-intensive. In recent years, Large Language Models (LLMs) such as ChatGPT 3.5 and generative models such as stable diffusion have emerged as powerful tools for comprehending natural language and generating digital contents such as text, code, images, and 3D objects. In this paper, we have explored how we can generate VR scenes from text by incorporating LLMs and various generative models into an automated system. To achieve this, we first identify the possible limitations of LLMs for an automated system and propose a systematic framework to mitigate them. Subsequently, we developed Text2VRScene, a VR scene generation system, based on our proposed framework with well-designed prompts. To validate the effectiveness of our proposed framework and the designed prompts, we carry out a series of test cases. The results show that the proposed framework contributes to improving the reliability of the system and the quality of the generated VR scenes. The results also illustrate the promising performance of the Text2VRScene in generating satisfying VR scenes with a clear theme regularized by our well-designed prompts. This paper ends with a discussion about the limitations of the current system and the potential of developing similar generation systems based on our framework. © 2024 IEEE.

This paper presents a thorough review of the evolution of Stacked Complementary Field-Effect transistors (CFET), including earlier work and recent demonstrations. The study outlines the challenges encountered during the fabrication process, such as insulator formation between active layers, deep ion implantation for bottom devices, selective contact between top and bottom transistors, and the layout of stacked transistors. Additionally, this paper explores the potential extension of the CFET technology and its integration with emerging device structures and advanced materials, such as 2D transistors. By analyzing these possibilities, this study sheds light on the potential for further advancements and applications of the CFET technology. © 2024 IEEE.

Neural Radiance Fields (NeRF) have shown impressive capabilities for photorealistic novel view synthesis when trained on dense inputs. However when trained on sparse inputs NeRF typically encounters issues of incorrect density or color predictions mainly due to insufficient coverage of the scene causing partial and sparse supervision thus leading to significant performance degradation. While existing works mainly consider ray-level consistency to construct 2D learning regularization based on rendered color depth or semantics on image planes in this paper we propose a novel approach that models 3D spatial field consistency to improve NeRF's performance with sparse inputs. Specifically we first adopt a voxel-based ray sampling strategy to ensure that the sampled rays intersect with a certain voxel in 3D space. We then randomly sample additional points within the voxel and apply a Transformer to infer the properties of other points on each ray which are then incorporated into the volume rendering. By backpropagating through the rendering loss we enhance the consistency among neighboring points. Additionally we propose to use a contrastive loss on the encoder output of the Transformer to further improve consistency within each voxel. Experiments demonstrate that our method yields significant improvement over different radiance fields in the sparse inputs setting and achieves comparable performance with current works. The project page for this paper is available at https://zhongyingji.github.io/CVT-xRF.

This paper adopts a methodology to assess and rank the severity of the most frequent disturbances, 3 types in particular, that occur in the Mexican Interconnected System (MIS) when wind power plants (WPPs) are integrated. This is done by combining three stability indices from which an overall performance measure can be derived. In this way, it is possible to assess the severity of the disturbances such as generation tripping, line tripping and disconnection of loads. The indices under consideration include measures such as maximum amplitude, speed variation, and the Lyapunov exponent. For ranking the severity, a general index value is used. To confirm the effectiveness and performance of the proposed methodology, this paper investigates the 190-bus and 46-generator MIS equipped with an 8% of wind generation to quantitatively evaluate the severity of disturbances, resulting in 66% of unstable disturbances which are associated with the disconnection of loads.

Estimating software testability can crucially assist software managers to optimize test budgets and software quality. In this paper, we propose a new approach that radically differs from the traditional approach of pursuing testability measurements based on software metrics, e.g., the size of the code or the complexity of the designs. Our approach exploits automatic test generation and mutation analysis to quantify the evidence about the relative hardness of developing effective test cases. In the paper, we elaborate on the intuitions and the methodological choices that underlie our proposal for estimating testability, introduce a technique and a prototype that allows for concretely estimating testability accordingly, and discuss our findings out of a set of experiments in which we compare the performance of our estimations both against and in combination with traditional software metrics. The results show that our testability estimates capture a complementary dimension of testability that can be synergistically combined with approaches based on software metrics to improve the accuracy of predictions.

This paper addresses some of the challenges of video forgery detection in multimedia forensics. It presents a solution that enhances video verification by utilizing the non-alterable features of blockchain technology and video hashing algorithms. The proposed approach is applicable in different application areas and can be used to increase video credibility, identify manipulations, and improve the storage process of tracking changes to video data. The paper describes our experiments and results of the proposed solution for video integrity preservation and verification, providing an alternative way to the quality assurance and security of video content in different industries.

With the widespread application of intelligent transportation systems and navigation software, traffic prediction should be modeled in finer granularity to facilitate lane-changing guidance and congestion mitigation. However, existing studies divide the road network into continuous segments which assumes different lanes share the same spatio-temporal patterns. This paper proposes a novel lightweight, attention-based, fully convolutional model, named the Spatio-Temporal Attention- Based Convolutional network (ST-ABC), where lane segments are treated as graph nodes and dynamically models the adjacent spatial dependencies using local attention graph convolution. The attention-based dilated convolutions can process longer sequence periods in parallel, and a global attention layer allows individual nodes to be associated with the global context. By setting a target window, it can further reduce unnecessary computations and improve the prediction effect for the targeted area. Further-more, the ST-ABC model facilitates the simultaneous integration of spatio-temporal information and relational distance metrics among lane segments, enriching the granularity of multi-scaled spatial prediction. Empirical evaluations conducted on two real-world datasets substantiate the augmented efficacy of the STABC model in comparison to established models, with a marked prominence in long-term prediction scenarios.

Owing to the nature of privacy protection, feder-ated recommender systems (FedRecs) have garnered increasing interest in the realm of on-device recommender systems. However, most existing FedRecs only allow participating clients to collaboratively train a recommendation model of the same public parameter size. Training a model of the same size for all clients can lead to suboptimal performance since clients possess varying resources. For example, clients with limited training data may prefer to train a smaller recommendation model to avoid excessive data consumption, while clients with sufficient data would benefit from a larger model to achieve higher recommendation accuracy. To address the above challenge, this paper introduces HeteFedRec, a novel FedRec framework that enables the assignment of personalized model sizes to partici-pants. Specifically, we present a heterogeneous recommendation model aggregation strategy, including a unified dual-task learning mechanism and a dimensional decorrelation regularization, to allow knowledge aggregation among recommender models of different sizes. Additionally, a relation-based ensemble knowledge distillation method is proposed to effectively distil knowledge from heterogeneous item embeddings. Extensive experiments conducted on three real-world recommendation datasets demonstrate the effectiveness and efficiency of HeteFedRec in training federated recommender systems under heterogeneous settings.

Recently, graph contrastive learning proposes to learn node representations from the unlabeled graph to alleviate the heavy reliance on node labels in graph neural networks (GNNs). The core idea is to generate diverse positive views and negative views according to local subgraphs. Then, GNNs take these views as supervised signals and train the model by maximizing the similarity between positive view pairs of each node and minimizing the similarity between positive and negative views. Regardless of the fruitful progress, existing graph contrastive learning approaches still suffer from low-efficiency, insufficient-expressivity, and unpreserved-locality issues. First, they train GNNs by all nodes, reducing the efficiency due to similar and redundant nodes. Second, they only use limited operations (e.g., edge deletion and feature masking) to generate positive views, thereby restricting their expressivity. Third, they uniformly delete edges and mask node features and may modify important edges and features, thereby damaging the important locality information of nodes. In this paper, we propose an efficient and expressive contrastive learning framework for GNNs, namely E 2 GCL. Specifically, given a limited node budget, we select a set of representative nodes instead of all nodes to accelerate the GNNs training. Besides, we use three general operations (edge deletion, edge addition, and feature perturbation) to generate expressive and locality-preserved positive views based on edge and feature importance. Extensive experiments on various real-world datasets demonstrate the superior effectiveness and efficiency of our proposed E 2 GCL.

Recently, continual learning (CL) has attracted much attention due to its widespread applications in the real world. Given a set of data sets sequentially, continual learning aims to achieve good performance on the new data sets while avoiding deterioration in performance on the old data sets. Despite the success, most CL models follow the supervised setting, which limits their potential in data scarcity cases. Thus, some pioneering works study unsupervised CL (UCL) to discuss what CL tricks suit the unsupervised setting. However, their advancements lack in-depth analysis of the characteristics of UCL, especially the lack of attention to the use of old data. We identify that using old data sets is essential for improving the UCL model performance while existing works ignore them. Unfortunately, given a limited data storage budget, it is a nontrivial task to select representative data and effectively replay them without label assistance. To further improve the UCL performance, we present a new method in this paper, named Effective Data Selection and Replay (EDSR) for UCL. Specifi-cally, we analyze that entropy can be an effective data selection metric, where representative data usually exhibit the highest entropy in the representation space. Then, to balance the model stability for old data and the plasticity for new data, we adopt a strategy of replaying those stored representative data with a noise-enhanced knowledge distillation process. The empirical study demonstrates the outstanding performance of EDSR on benchmark computer vision data sets. Especially, EDSR shows strong resistance to forgetting old data knowledge while maintaining high accuracy. The implementation is publicly available at https://github.com/LeeJarvis996/edsr_project/tree/main/EDSR.

With the popularization of mobile applications and the timely acquisition of fresh data, real-time clustering and its evolution analysis have become the primary operations for data processing and knowledge discovery. Such continuous queries on massive objects are computation-intensive tasks in dynamic scenarios. However, existing clustering techniques are incompetent to achieve decent performance when computation-intensive operations frequently occur in streaming scenarios, which is caused by two challenges: (i) uncertainty of the clustering frequency; (ii) unpredictable distribution evolution. Hence, it is critical to find a lightweight model that can cluster the high-speed dynamic instances while exploiting the evolution amid different clustering results. This paper focuses on the problem of real-time clustering on streaming data in computation-intensive and high-dynamics tasks, through a framework Ocean, consisting of the Online clustering algorithm and evolution analysis. Particularly, the framework conceives a flexible composite window to augment the knowledge mining, achieving a proper real-time response in various scenarios. The evolution analysis supports full life-cycle detection, improving the adaptability to dynamic concept drifts and multiple patterns. Inspired by the grid partition strategy, this framework adopts grid feature vectors to capture the significant changes in streaming data. Furthermore, we propose an optimization that removes sparse grids timely and performs the online clustering adaptively for space and time efficiency. It is proven to be effective both theoretically and experimentally. This strategy enables real-time clustering for dynamic streaming data without degrading the clustering quality or increasing the computation cost. Experiments on real datasets and synthetic datasets verify the accuracy and effectiveness of Ocean compared to the state-of-the-art approaches, as well as the superior ability to perform clustering in a real-time

Sequential Recommendation (SR) has received increasing attention due to its ability to capture user dynamic preferences. Recently, Contrastive Learning (CL) provides an effective approach for sequential recommendation by learning invariance from different views of an input. However, most existing data or model augmentation methods may destroy semantic sequential interaction characteristics and often rely on the hand-crafted property of their contrastive view-generation strategies. In this paper, we propose a Meta-optimized Seq2Seq Generator and Contrastive Learning (Meta-SGCL) for sequential recommendation, which applies the meta-optimized two-step training strategy to adaptive generate contrastive views. Specifically, Meta-SGCL first introduces a simple yet effective augmentation method called Sequence-to-Sequence (Seq2Seq) generator, which treats the Variational AutoEncoders (VAE) as the view generator and can constitute contrastive views while preserving the original sequence's semantics. Next, the model employs a meta-optimized two-step training strategy, which aims to adaptively generate contrastive views without relying on manually designed view-generation techniques. Finally, we evaluate our proposed method Meta-SGCL using three public real-world datasets. Compared with the state-of-the-art methods, our experimental results demonstrate the effectiveness of our model and the code is available. https.//anonymous.4open.science/status/Meta-SGCL-05B5

The Constrained Shortest Path (CSP) problem seeks to identify the shortest path between two vertices in a road network while adhering to a specific constraint on another criterion. Solving the CSP problem frequently entails navigating the two-criteria skyline path problem, which incurs a substantial computational expense in large road networks. The primary challenge lies in handling a vast quantity of partial skyline paths, which often hinders index-based solutions from accurately determining the skyline paths. This paper introduces a-FHL, a practical approximation method designed to circumvent the costly skyline path search and hasten computation on skyline path indexing. a-FHL uses tree decomposition to hierarchically assign approximation ratios, thereby facilitating effective pruning within the labelling index. Moreover, we devise various strategies to allocate approximation ratios and an efficient approximation concatenation method to respond to the approximate CSP queries via the a-FHL index. Our method culminates in swift index construction and efficient query response. Comprehensive exper-iments conducted on real-world road networks substantiate the superiority of our approach over contemporary solutions

Trust plays an essential role in an individual's decision-making. Traditional trust prediction models rely on pairwise correlations to infer potential relationships between users. However, in the real world, interactions between users are usually complicated rather than pairwise only. Hypergraphs offer a flexible approach to modeling these complex high-order correlations (not just pairwise connections), since hypergraphs can leverage hyperedeges to link more than two nodes. However, most hypergraph-based methods are generic and cannot be well applied to the trust prediction task. In this paper, we propose an Adaptive Hypergraph Network for Trust Prediction (AHNTP), a novel approach that improves trust prediction accuracy by using higher-order correlations. AHNTP utilizes Motif-based PageRank to capture high-order social influence information. In addition, it constructs hypergroups from both node-level and structure-level attributes to incorporate complex correlation information. Furthermore, AHNTP leverages adaptive hypergraph Graph Convolutional Network (GCN) layers and multilayer perceptrons (MLPs) to generate comprehensive user embeddings, facilitating trust relationship prediction. To enhance model generalization and robustness, we introduce a novel supervised contrastive learning loss for optimization. Extensive experiments demonstrate the superiority of our model over the state-of-the-art approaches in terms of trust prediction accuracy.

This paper considers cell-free multiple-input multiple-output (MIMO) systems with multi-antenna access points (APs) and multi-antenna users in constrained spaces. Two main effects emerge in such a space-constrained design: spatial correlation and mutual coupling. We analytically study the resulting performance with conjugate beamforming and reducing inter-antenna distance. We derive the closed-form expression for the spectral efficiency (SE) as a function of the spatial correlation and mutual coupling matrices. Additionally, we consider a practical power consumption model to investigate the energy efficiency (EE) with power control coefficients. The obtained ergodic and analytical results will show essential insights into the system performance. The decreasing inter-antenna distance introduces performance loss, while an appropriate number of APs could balance the SE and EE trade-off. The number of users improves the SE performance, while an optimal number of antennas per user can be selected to achieve maximum SE performance. Moreover, more users will introduce a smaller optimal number of antennas per user.

The maximum inner product search (MIPS) problem in high-dimensional vector spaces has various applications, primarily driven by the success of deep neural network-based embedding models. Existing MIPS methods designed for dense vectors using approximate techniques like locality-sensitive hashing (LSH) have been well studied, but they are not efficient and effective for searching sparse vectors due to the near-orthogonality among the sparse vectors. The solutions to MIPS over sparse vectors rely heavily on inverted lists, resulting in poor query efficiency, particularly when dealing with large-scale sparse datasets. In this paper, we introduce SOSIA, a novel framework specifically tailored to address these limitations. To handle sparsity, we propose the SOS transformation, which converts sparse vectors into a binary space while providing an unbiased estimator of the inner product between any two vectors. Additionally, we develop a minHash-based index to enhance query efficiency. We provide a theoretical analysis on the query quality of SOSIA and present extensive experiments on real-world sparse datasets to validate its effectiveness. The experimental results demonstrate its superior performance in terms of query efficiency and accuracy compared to existing methods.

Deep reinforcement learning (RL) has emerged as a promising approach for portfolio management due to its ability to make sequential decisions. However, applying RL techniques to this domain is still challenging due to the non-stationary nature of financial markets. Existing RL-based solutions fail to consider the intrinsic causes behind this non-stationary, which primarily stem from the involvement of diverse traders with distinct investment horizons and their varied investment strategies. In this paper, we tackle the non-stationary problem by examining its intrinsic causes and propose cross-insight trader, a novel two-step RL-based approach that integrates multiple trading policies with different investment horizons to adapt to the changing market conditions. In the first step, we learn multiple horizon-specific policies by providing each policy with tailored information specific to its investment horizon. This allows each policy to recognize dynamic patterns within its respective horizon and make insightful pre-decisions. In the second step, we learn a cross-insight policy to make the final trade decision by considering the investment pre-decisions made by multiple horizon-specific policies in the first step. To enable effective learning of two types of policies, our approach employs a centralized critic to evaluate the actions performed by both horizon-specific and cross-insight policies. By incorporating multiple insights from different investment horizons into the decision-making process, our approach enhances its adaptability to changing market conditions. Experimental results conducted on three stock markets demonstrate the superiority of our framework.

In this paper we introduce GS-SLAM that first utilizes 3D Gaussian representation in the Simultaneous Localization and Mapping (SLAM) system. It facilitates a better balance between efficiency and accuracy. Compared to recent SLAM methods employing neural implicit representations our method utilizes a real-time differentiable splatting rendering pipeline that offers significant speedup to map optimization and RGB-D rendering. Specifically we propose an adaptive expansion strategy that adds new or deletes noisy 3D Gaussians in order to efficiently reconstruct new observed scene geometry and improve the mapping of previously observed areas. This strategy is essential to extend 3D Gaussian representation to reconstruct the whole scene rather than synthesize a static object in existing methods. Moreover in the pose tracking process an effective coarse-to-fine technique is designed to select reliable 3D Gaussian representations to optimize camera pose resulting in runtime reduction and robust estimation. Our method achieves competitive performance compared with existing state-of-the-art real-time methods on the Replica TUM-RGBD datasets. Project page: \href https://gs-slam.github.io/ https://gs-slam.github.io/ .

Pervasive Augmented Reality (AR) requires accurate pose registration of the device in real-time at a neighbourhood-to-city scale. At such a scale, most pose registration techniques suffer from exponential computational and storage costs and a significant data collection burden. This paper introduces AnchorLoc, a framework that relies on visual anchors (stable and highly recognisable visual elements in a scene) to perform fast and accurate pose registration. Anchorloc automatically identifies these anchors from large image sequences to optimise the search space in later image retrieval and pose registration. As such, it significantly improves the computational efficiency of existing hierarchical localisation pipelines without compromising accuracy. We collect a large-scale localisation dataset consisting of image sequences and 3D reconstruction of a university campus. AnchorLoc reduces localisation runtime by 83% on our campus dataset and 69% on the Cambridge Landmarks dataset without significantly increasing mean pose estimation errors. It is also more accurate and faster than SLD, a localisation algorithm that takes a comparable approach at the keypoint level. This work informs the development of more efficient pervasive AR applications that rely on both absolute and relative camera pose tracking on image sequences. © 2024 IEEE.

Existing open-vocabulary object detectors typically require a predefined set of categories from users significantly confining their application scenarios. In this paper we introduce DetCLIPv3 a high-performing detector that excels not only at both open-vocabulary object detection but also generating hierarchical labels for detected objects. DetCLIPv3 is characterized by three core designs: 1. Versatile model architecture: we derive a robust open-set detection framework which is further empowered with generation ability via the integration of a caption head. 2. High information density data: we develop an auto-annotation pipeline leveraging visual large language model to refine captions for large-scale image-text pairs providing rich multi-granular object labels to enhance the training. 3. Efficient training strategy: we employ a pre-training stage with low-resolution inputs that enables the object captioner to efficiently learn a broad spectrum of visual concepts from extensive image-text paired data. This is followed by a fine-tuning stage that leverages a small number of high-resolution samples to further enhance detection performance. With these effective designs DetCLIPv3 demonstrates superior open-vocabulary detection performance e.g. our Swin-T backbone model achieves a notable 47.0 zero-shot fixed AP on the LVIS minival benchmark outperforming GLIPv2 GroundingDINO and DetCLIPv2 by 18.0/19.6/6.6 AP respectively. DetCLIPv3 also achieves a state-of-the-art 19.7 AP in dense captioning task on VG dataset showcasing its strong generative capability.

Annotation through crowdsourcing draws incremental attention, which relies on an effective selection scheme given a pool of workers. Existing methods propose to select workers based on their performance on tasks with ground truth, while two important points are missed. 1) The historical performances of workers in other tasks. In real-world scenarios, workers need to solve a new task whose correlation with previous tasks is not well-known before the training, which is called cross-domain. 2) The dynamic worker performance as workers will learn from the ground truth. In this paper, we consider both factors in designing an allocation scheme named cross-domain-aware worker selection with training approach. Our approach proposes two estimation modules to both statistically analyze the cross-domain correlation and simulate the learning gain of workers dynamically. A framework with a theoretical analysis of the worker elimination process is given. To validate the effectiveness of our methods, we collect two novel real-world datasets and generate synthetic datasets. The experiment results show that our method outperforms the baselines on both real-world and synthetic datasets.

The increasing demand for computational photography and imaging on mobile platforms has led to the widespread development and integration of advanced image sensors with novel algorithms in camera systems. However, the scarcity of high-quality data for research and the rare opportunity for in-depth exchange of views from industry and academia constrain the development of mobile intelligent photography and imaging (MIPI). Building on the achievements of the previous MIPI Workshops held at ECCV 2022 and CVPR 2023, we introduce our third MIPI challenge including three tracks focusing on novel image sensors and imaging algorithms. In this paper, we summarize and review the Few-shot RAW Image Denoising track on MIPI 2024. In total, 165 participants were successfully registered, and 7 teams submitted results in the final testing phase. The developed solutions in this challenge achieved state-of-the-art performance on Few-shot RAW Image Denoising. More details of this challenge and the link to the dataset can be found at https://mipichallenge.org/MIPI2024.

Much progress has been made in dynamic scene reconstruction by neural rendering techniques. Even if existing methods based on Neural Radiance Field (NeRF) achieve marvelous fidelity, they suffer from either complex camera settings, dense training timestamps, or limited ability to inter- and extrapolate. In this paper, a novel paradigm is presented to address these issues. Our method, called Dynamic Ode NErf (DONE), comprises two main modules. The first one trains a deformable neural representation in a static scene to store neural features. The second module establishes a deterministic neural ordinary differential equation to model the dynamics of the scene object. We also present a new 360-degree dynamic dataset for the research of entire 360-degree dynamic scene reconstruction. Experiments visually and quantitatively illustrate the effectiveness of the proposed model. © 2024 IEEE.

Time-varying graphical LASSO (TVGL) aims to infer a sequence of graphs from time series data and has been widely used in many statistical inference problems. The existing algorithms usually suffer from high computational cost when solving large-scale TVGL problems. In this paper, we develop an efficient and scalable hierarchical block coordinate descent (HBCD) method for solving TVGL with smooth temporal difference prior. The proposed HBCD method contains both outer-loop and inner-loop BCD iterations. The outer loops seperate the original TVGL problem into a sequence of subproblems, which are variants of the static graphical LASSO problems. Then, we propose an efficient BCD method to solve the inner-loop subproblems. We provide theoretical analysis that indicates the linear convergence of our proposed method. Furthermore, numerical experiments on both synthetic and real datasets show that our method significantly outperforms the state-of-the-art algorithm in terms of both the required iterations and CPU time to reach the target precision. © 2024 IEEE.

This paper uses multimodal haptic feedback to enhance users' body ownership in virtual reality through wearable devices. In this case, the human is transformed into a bird, which belongs to the beyondreal transformations category in virtual reality interactions. For body transformation, wearable retractable straps can help people mimic the movement mechanism of avian bodies; for space transformation, the inflatable cushions and blowers can simulate the air resistance and lift, oxygen deprivation, and temperature decrease during the take-off process of avian avatars. The system aims to establish a realistic fidelity of the haptic feedback to enhance the user's body ownership. © 2024 IEEE.

Large language models (LLMs) offer impressive performance across diverse fields, but their increasing complexity raises both design costs and the need for specialized expertise. These challenges are intensified for Neural Architecture Search (NAS) methods reliant on weight-sharing techniques. This paper introduces GNAS, a new NAS method that boosts the search process with the aid of LLMs for efficient model discovery. With insights from existing architectures, GNAS swiftly identifies superior models that can adapt to changing resource constraints. We provide a mathematical framework to facilitate the transfer of knowledge across different model sizes, thereby improving search efficiency. Our experiments conducted on ImageNet, NAS-Bench-Macro, and Channel-Bench-Macro confirm the effectiveness of GNAS across both CNN and Transformer architectures. © 2024 IEEE.

The purpose of this paper is to use an advanced spiking neural network algorithm, and accelerate the algorithm with a variety of acceleration methods based on the high-level synthesis platform. Its target is to solve the problem of insufficient performance of the traditional computing platform in real-time electrocardiogram data processing. The experiment shows that the performance is improved significantly, which proves that the proposed method is effective in real-time data processing. © 2024 IEEE.

Augmented Reality (AR) teleconferencing aims to enable remotely separated users to meet with each other in their own physical spaces as if they are face-to-face. Among all solutions, the video-avatar-based approach has the advantage of balancing fidelity and the sense of co-presence using easy-to-setup devices, including only a camera and an AR Head-Mounted Display (HMD). However, non-verbal cues indicating 'who is looking at whom' are always lost or misdelivered in multiparty teleconferencing experiences. To make users aware of such non-verbal cues, existing solutions explore screen-based visualizations, incorporate additional hardware, or alter to use a virtual avatar representation. However, they lack immersion, are less feasible for everyday usage due to complex installations, or lose the fidelity of remote users' authentic appearances. In this paper, we decompose such attention awareness into the awareness of being looked at and the awareness of attention between other users and address them in a decoupled process. Specifically, through a user study, we first find an unobtrusive and reasonable layout 'Attention Circle' to retarget a looker's head gaze to the one being looked at. We then conduct the second user study to find an effective and intuitive 'rotatable 2.5D video avatar with attention thumbnail' visualization to aid users in being aware of other users' attention. With the design choice distilled from the studies, we implement A3RT, a proof-of-concept prototype system that empowers attention-aware 2.5D-video-avatar-based multiparty AR teleconferencing in an easy, everyday setup. Ablation and usability studies on the prototype verify the effectiveness of our proposed components and the full system.

Ambient radio frequency (RF) powered Internet of Things (IoT) is a promising batteryless system for collecting environmental information for data exchange among devices. Ambient RF energy harvesting technology is most suitable for use in places that are hidden from light sources and are in difficult to reach places for battery replacement. In this paper, a demonstration of IoT sensors powered by ambient RF energy harvesting is presented. The system we have developed provides all key components from antenna, rectifier, power management unit (PMU), microcontroller unit (MCU) to sensing unit. In addition a novel 3-port antenna element for the systems is also described. The experimental results confirm that ambient RF signals can be regarded as an alternative to power supply in the next generation of wireless communications. © 2024 IEEE.

In this paper, we introduce a novel micro-transfer printing method for photoresist transfer. This approach utilizes the thermo-responsive adhesion modulation properties of a shape memory polymer (SMP) stamp, to facilitate the efficient transfer of patterned photoresist from a donor substrate to various unconventional receiver, such as water-soluble, stretchable, and curved surfaces, which typically present challenges in direct photolithography processes. Our study successfully demonstrates the effective transfer of various photoresists (PR), including AZ 5214E and SU-8 2010, onto diverse substrates, such as polyvinyl alcohol (PVA), silicone and curved glass substrates, highlighting the versatility of the technology. This technology paves the way for the development of advanced wearable electronics, biomedical sensors, and implantable medical devices, representing a progression in the high-resolution microfabrication of innovative electronics for a wide range of applications. © 2024 IEEE.

Spiking neural networks (SNNs) have ultra-low energy consumption and high biological plausibility due to their binary and bio-driven nature compared with artificial neural networks (ANNs). While previous research has primarily focused on enhancing the performance of SNNs in classification tasks, the generative potential of SNNs remains relatively unexplored. In our paper, we put forward Spiking Denoising Diffusion Probabilistic Models (SDDPM), a new class of SNN-based generative models that achieve high sample quality. To fully exploit the energy efficiency of SNNs, we propose a purely Spiking U-Net architecture, which achieves comparable performance to its ANN counterpart using only 4 time steps, resulting in significantly reduced energy consumption. Extensive experimental results reveal that our approach achieves state-of-the-art on the generative tasks and substantially outperforms other SNN-based generative models, achieving up to 12× and 6× improvement on the CIFAR-10 and the CelebA datasets, respectively. Moreover, we propose a threshold-guided strategy that can further improve the performances by 2.69% in a training-free manner. The SDDPM symbolizes a significant advancement in the field of SNN generation, injecting new perspectives and potential avenues of exploration. Our code is available at https://github.com/AndyCao1125/SDDPM. © 2024 IEEE.

Virtual reality (VR) is an immersive medium that offers users a unique opportunity to experience a digital environment realistically. As the demand for VR content continues to grow, the importance of effective VR editing techniques becomes increasingly apparent. This paper is a pioneering work investigating the effects of jump cuts on the viewer's sense of presence, viewing experience, and edit quality in cinematic VR. Specifically, this work focuses on using the 30-degree and 180-degree rules in VR editing to minimize the adverse effects of jump cuts. We conducted a user study with thirteen participants, who watched nine different VR edits and completed a survey for each edited video. Our results indicate that employing the 30-degree and 180-degree rules in VR editing can significantly improve the sense of presence, viewing experience, and edit quality while mitigating the negative effects of jump cuts. We provide valuable insights for VR content creators and editors to achieve more effective and immersive VR experiences. © 2024 IEEE.

Multi-task learning for dense prediction has emerged as a pivotal area in computer vision enabling simultaneous processing of diverse yet interrelated pixel-wise prediction tasks. However the substantial computational demands of state-of-the-art (SoTA) models often limit their widespread deployment. This paper addresses this challenge by introducing network binarization to compress resource-intensive multi-task dense predictors. Specifically our goal is to significantly accelerate multi-task dense prediction models via Binary Neural Networks (BNNs) while maintaining and even improving model performance at the same time. To reach this goal we propose a Binary Multi-task Dense Predictor Bi-MTDP and several variants of \bimtdp in which a multi-task dense predictor is constructed via specified binarized modules. Our systematical analysis of this predictor reveals that performance drop from binarization is primarily caused by severe information degradation. To address this issue we introduce a deep information bottleneck layer that enforces representations for downstream tasks satisfying Gaussian distribution in forward propagation. Moreover we introduce a knowledge distillation mechanism to correct the direction of information flow in backward propagation. Intriguingly one variant of Bi-MTDP outperforms full-precision (FP) multi-task dense prediction SoTAs ARTC (CNN-based) and InvPT (ViT-based). This result indicates that Bi-MTDP is not merely a naive trade-off between performance and efficiency but is rather a benefit of the redundant information flow thanks to the multi-task architecture.

Orthogonal time frequency space (OTFS) modulation is strongly considered as a promising solution for high-mobility communications. In contrast to conventional modulation techniques, wherein information symbols are multiplexed in a one-dimensional time or frequency domain, OTFS employs a two-dimensional modulation scheme by multiplexing information symbols in the delay-Doppler domain. This paper presents a comprehensive survey of OTFS. It starts with an overview of OTFS, its advantages over conventional air interface techniques, general block diagrams, and implementations. Subsequently, the paper explores the potential integration of multiple-input multiple-output and OTFS techniques. The paper further discusses the feasibility of integrating OTFS into multiple access techniques as a solution for maintaining acceptable performance in high-mobility scenarios. Then, widespread applications of OTFS in satellite communications are highlighted. Also, the potential utilization of OTFS modulation in integrated sensing and communications paradigm is thoroughly treated. In addition, the survey covers further applications of OTFS in deep learning, index modulation, underwater acoustic, and unmanned aerial vehicle communications. The paper concludes by pointing out numerous challenging and promising directions for future OTFS research.

In real-world scenarios, gas sensor responses to mixtures of different compositions can be costly to determine a-priori, posing difficulties in identifying the presence of target analytes. In this paper, we propose the use of graph convolutional networks (GCN) to handle gas mixtures with few labelled data. We transform sensor responses into a graph structure using manifold learning and clustering, and then apply GCN for semisupervised node classification. Our approach does not require extensive training data of gas mixtures like many competing approaches, but it outperforms classical semi-supervised learning methods and achieves classification accuracy exceeding 88.5% and over 0.85 Cohen's kappa score given only 5% labelled data for training. This result demonstrates the potential towards realistic gas identification when varied mixtures are present., This work is supported by the project SP13 'Monitoring of airflow and airborne particles, to provide early warning of irrespirable atmospheric conditions' under the academic program Sustainable Underground Mining (SUM), jointly financed by LKAB and the Swedish Energy Agency.

To effectively monitor a network and verify its performance, it is essential to quickly detect sudden changes in its state, even when the form of such a change is initially unknown. While classical quickest change detection methods are potentially useful, they rely on probing the network state periodically, which in turn, may induce high measurement costs. In this paper, we extend existing frameworks in quickest change detection to allow for both adaptive measurement periods and unknown post-change measurement distribution. In our extended framework, the agent decides both when to raise an alarm and when to take the next measurement (if any), maintaining a trade-off between detection delay, false alarm rate and measurement costs. We evaluate both classical methods with periodic measurements as well as our adaptive scheme, called RoME-QCD (Robust and Measurement Efficient Quickest Change Detection). We demonstrate the latter's superiority analytically and verify this observation via numerical experiments, both using one-way delay data from a 5G testbed and synthetically generated data., QC 20240812

Citizen trust is a central part of an ethical AI ecosystem, and a key part of upcoming AI legislation across the world. However, pockets of society have little awareness of what is meant by AI or its current use by organizations. Therefore, a first step towards building citizen trust is to raise general awareness of what is meant by AI technologies, how data is captured and used in decision making, and existing citizens’ rights to question organizations about the use of their data in automated decision making. This paper describes a mechanism to reach different groups of publics through a Community AI Roadshow. The motivation was to develop a way to reach and engage with traditionally marginalized communities and develop a common language and understanding around AI. An evaluation showed that understanding of AI increased by 33% following the roadshow. A resulting set of recommendations for AI researchers engaging with marginalized communities is given. The methodology presented in this paper has since been adapted for other groups of publics, such as local governmental organizations in the UK.

This paper proposes the application of logic locking over TFHE to protect both user data and algorithms, such as input user data and models in machine learning inference applications. With the proposed secure computation protocol algorithm evaluation can be performed distributively on honest-but-curious user computers while keeping the algorithm secure. To achieve this, we combine conventional logic locking for untrusted foundries with TFHE to enable secure computation. By encrypting the logic locking key using TFHE, the key is secured with the degree of TFHE. We implemented the proposed secure protocols for combinational logic neural networks and decision trees using LUT-based obfuscation. Regarding the security analysis, we subjected them to the SAT attack and evaluated their resistance based on the execution time. We successfully configured the proposed secure protocol to be resistant to the SAT attack in all machine learning benchmarks. Also, the experimental result shows that the proposed secure computation involved almost no TFHE runtime overhead in a test case with thousands of gates.

As the uplink sensing has the advantage of easy implementation it attracts great attention in integrated sensing and communication ISAC system This paper presents an uplink ISAC system based on multi input multi output orthogonal frequency division multiplexing MIMO OFDM technology The mutual information MI is introduced as a unified metric to evaluate the performance of communication and sensing In this paper firstly the upper and lower bounds of communication and sensing MI are derived in details based on the interaction between communication and sensing And the ISAC waveform is optimized by maximizing the weighted sum of sensing and communication MI The Monte Carlo simulation results show that compared with other waveform optimization schemes the proposed ISAC scheme has the best overall performance

This paper presents an overview of the advances in reconfigurable Ygai Uda antenna for smart wireless communications systems In particularly it reports the development of planar Yagi Uda antenna with frequency and pattern reconfiguration Furthermore this paper presents a recent design on a circular array of Yagi Uda monopoles with a full 360 azimuthal beam scanning