Estimating the frequency of items on the high-volume, fast data stream has been extensively studied in many areas, such as database and network measurement. Traditional sketches provide only coarse estimates under strict memory constraints. Although some learning-augmented methods have emerged recently, they typically rely on offline training with real frequencies or/and labels, which are often unavailable. Moreover, these methods suffer from slow update speeds, limiting their suitability for real-time processing despite offering only marginal accuracy improvements. To overcome these challenges, we propose UCL-sketch, a practical learning-based paradigm for per-key frequency estimation. Our design introduces two key innovations: (i) an online training mechanism based on equivalent learning that requires no ground truth (GT), and (ii) a highly scalable architecture leveraging logically structured estimation buckets to scale to real-world data stream. The UCL-sketch, which utilizes compressive sensing (CS), converges to an estimator that provably yields an error bound far lower than that of prior works, without sacrificing the speed of processing. Extensive experiments on both real-world and synthetic datasets demonstrate that our approach outperforms previously proposed approaches regarding per-key accuracy and distribution. Notably, under extremely tight memory budgets, its quality almost matches that of an (infeasible) omniscient oracle. Moreover, compared to the existing equation-based sketch, UCL-sketch achieves an average decoding speedup of nearly 500 times.

With the rapid development of application containerization, Kubernetes has become the mainstream container orchestration system. However, the Kubernetes default scheduler uses static scheduling strategies. It did not take into account the possibility of inconsistency between the resource utilization of Pod during runtime and its own resource request claim, leading to a large amount of resource fragmentation on actual cluster nodes. In specific cases, unlike normal workload fluctuations, we have observed “surge” phenomena where an application usually only uses a small amount of resources for a long period of time but exhausts all resources in a short period of time. This could lead to service disruptions. For this application scenario, we have proposed a Kubernetes scheduling algorithm, KS-surge. It not only ensures the load balancing of system resources but also can effectively handle sudden situations such as a surge in task requests. The experimental results show that, compared to the default Kubernetes resource scheduler, our proposed algorithm improves the load balancing of the clusters while meeting the requirements of applications.

Aiming at the problem of task offloading in multiaccess edge computing (MEC) scenarios, this paper proposes a meta-reinforcement learning (Meta-RL)-based computational task offloading method. The algorithm adopts a two-layer architecture: the inner layer models the task offloading process as a Markov Decision Process (MDP), designs a reward function based on task latency and energy consumption, and designs a task offloading algorithm based on Proximal Policy Optimization (PPO) to make offloading decisions for each task in a single scenario and optimize the offloading performance within the scenario. The outer layer introduces the Meta-RL mechanism to optimize the initial parameters of the inner-layer neural network and learns multiple MDPs based on gradient descent to generate neural network parameters that can be applied to the intelligence of each scenario, so that the proposed algorithm can adapt to the offloading scenarios quickly. The proposed algorithm can quickly adapt to each offloading scenario. Simulation results show that the proposed algorithm improves the average cost by 14.7 % and 20.51 % compared with PPO and DDPG.

Diffusion models have exhibited state-of-the-art performance in generative tasks across various domains. A few recent works leveraged the powerful modeling ability of the diffusion model to time-series forecasting, leading to a significant breakthrough. However, all these works perform the forecasting through incorporating the historical time-series conditions into the backward denoising. This causes the diffusion model to lose the essential consistency between forward and backward processes, thereby limiting the precision of the inference. In this paper, we propose a novel Dual Trajectory Revised Diffusion Model (TimeDTR) for time-series forecasting, which leverages an unconventional conditioning strategy to incorporate the historical information into both forward and backward trajectories in the diffusion model. Experimental results on six real-world datasets demonstrate that TimeDTR takes a big step forward from the state-of-the-art in time-series forecasting, especially in the long-term forecasting tasks, in terms of forecasting accuracy. The codes of the experiments with datasets and our algorithms are available at https://github.com/hhzzlll/TimeDTR.

The decoder is a key component in deep networks for the semantic segmentation of street views. The existing methods rely on the limited receptive field for feature extraction without considering the contextual information, which could lead to errors in understanding complex scenes. Moreover, a balance of contextual information and computational cost must be considered to meet the needs of real-world applications. To address these problems, we introduce a Spatial Plaid Attention Decoder network, which uses a lightweight decoder with Spatial Plaid Attention to perform highly efficient operations for semantic segmentation. With approximately 4 million parameters (9.75% of the UPerNet), our decoder achieves state-of-the-art performance on public datasets such as Cityscapes and ADE20K, with 84.84% and 54.0% mIoU, respectively. In addition, our method reduces the total Flops by 34.95% and 32.85%, respectively. We demonstrate how contextual information helps the network in object recognition and how object features and contextual features contribute to the scene segmentation and recognition.

Due to network operation and maintenance relying heavily on network traffic monitoring, traffic matrix analysis has been one of the most crucial issues for network management related tasks. However, it is challenging to reliably obtain the precise measurement in computer networks because of the high measurement cost, and the unavoidable transmission loss. Although some methods proposed in recent years allowed estimating network traffic from partial flow-level or link-level measurements, they often perform poorly for traffic matrix estimation nowadays. Despite strong assumptions like low-rank structure and the prior distribution, existing techniques are usually task-specific and tend to be significantly worse as modern network communication is extremely complicated and dynamic. To address the dilemma, this paper proposed a diffusion-based traffic matrix analysis framework named Diffusion-TM, which leverages problem-agnostic diffusion to notably elevate the estimation performance in both traffic distribution and accuracy. The novel framework not only takes advantage of the powerful generative ability of diffusion models to produce realistic network traffic, but also leverages the denoising process to unbiasedly estimate all end-to-end traffic in a plug-and-play manner under theoretical guarantee. Moreover, taking into account that compiling an intact traffic dataset is usually infeasible, we also propose a two-stage training scheme to make our framework be insensitive to missing values in the dataset. With extensive experiments with real-world datasets, we illustrate the effectiveness of Diffusion-TM on several tasks. Moreover, the results also demonstrate that our method can obtain promising results even with 5% known values left in the datasets.

As part of airport behavior decisions, the accurate prediction of airport delay is highly significant in optimizing flight takeoff and landing sequences. However, the combination of various influencing factors affects airport delay prediction strongly, which would bring severe challenges in prediction. This paper introduces the sequence-to-sequence network and proposes a multi-step regression prediction method for the airport delay (DA-BILSTM) to accurately predict the airport delay. Rather than only considering a single kind of airport delay influencing factors, we design an attention fusion network for learning the sequence and condition correlation features adaptively. Moreover, the Bayesian optimization algorithm is introduced to optimize DA-BILSTM’s hyperparameters. The method is applied individually to two datasets for predicting the airport’s delays. The experiment results show that the prediction performance of DA-BILSTM is better than many state-of-the-art methods including the autoregressive integrated moving average model (ARIMA), long short-term memory (LSTM), gated recurrent unit(GRU), CNN-BILSTM, and TS-LSTM. When using DA-BILSTM in the two datasets, the average MAE of airport delay prediction in the next 5 hours is about 10 minutes, and the average RMSE is 20 minutes.

The vehicular edge computing (VEC) is a new paradigm that allows vehicles to offload computational tasks to base stations (BSs) with edge servers for computing. In general, the VEC paradigm uses the 5G for wireless communications, where the massive multi-input multi-output (MIMO) technique will be used. However, considering in the VEC environment with many vehicles, the energy consumption of BS may be very large. In this paper, we study the energy optimization problem for the massive MIMO-based VEC network. Aiming at reducing the relevant BS energy consumption, we first propose a joint optimization problem of computation resource allocation, beam allocation and vehicle grouping scheme. Since the original problem is hard to be solved directly, we try to split the original problem into two subproblems and then design a heuristic algorithm to solve them. Simulation results show that our proposed algorithm efficiently reduces the BS energy consumption compared to other schemes.

Abstract Software project defect prediction can help developers allocate debugging resources. Existing software defect prediction models are usually based on machine learning methods, especially deep learning. Deep learning‐based methods tend to build end‐to‐end models that directly use source code‐based abstract syntax trees (ASTs) as input. They do not pay enough attention to the front‐end data representation. In this paper, we propose a new framework to represent source code called multiperspective tree embedding (MPT‐embedding), which is an unsupervised representation learning method. MPT‐embedding parses the nodes of ASTs from multiple perspectives and encodes the structural information of a tree into a vector sequence. Experiments on both cross‐project defect prediction (CPDP) and within‐project defect prediction (WPDP) show that, on average, MPT‐embedding provides improvements over the state‐of‐the‐art method.

In general, edge devices and edge servers in edge computing environment communicate with each other by wireless network, which put forward a high requirement for end-to-end wireless communication performance. In this paper, we propose an optimal strategy by combining the coherent beamforming (CB) technique and the successive interference cancellation (SIC) technique for improving the performance of the edge device communications. CB technique can be used for expanding the transmitter's transmitting range, while SIC technique can be used for improving the receiver's receiving ability. However, when these two techniques are used jointly, interference will occur between transmitters and receivers, which makes the CB-SIC strategy hard to be designed. We first give the mathematical model based on CB-SIC and show it is difficult to solve directly. Then, we design a heuristic algorithm called time slot loop allocation (TSLA) algorithm. TSLA is based on greedy strategy to obtain an approximate optimal solution. By using TSLA, the whole scheduling time will be divided into many time slots. In each time slot, we try to make as many edge devices as possible to transmit data to the server. These can increase the overall data throughput. In simulation, we compare CB-SIC wireless network with CB only, SIC only, and traditional multi-hop network. Simulation results show that the TSLA algorithm can improve the end-to-end communication performance in edge computing environment.

Mobile Edge Computing (MEC) can significantly save the energy consumption of small-cell base stations(SBSs) by using Dynamic Voltage Scaling (DVS) technology. In this paper, we study how to reduce energy consumption of SBSs by using DVS technology. We propose a scheme to divide base station groups in wireless MEC network, and SBSs in the same base station group collaboratively process tasks. For cooperating tasks in a base station group, we propose a task-offloading strategy, which can effectively reduce the energy consumption of the base station groups. We use Task-offloading Decision Algorithm (TDA) to decide the number of task fragments allocated to each SBSs in the base station group. For processing task fragments in the task queue of SBSs, we propose a computing resource allocation scheme, and we can get the processing time of every task fragments in the task queue by Task Fragments Processing Algorithm (TFPA). Experimental results show that the task-offloading strategy and computing source allocation scheme can reduce the energy consumption by 30%-40% in a cellular network composed of 100 base stations.

Predicting short-term urban traffic flow is a non- trivial task, for an intelligent transportation system could greatly facilitate urban transportation infrastructure construction and enhances the efficiency of traffic control. Unfortunately, urban traffic flow is influenced by numerous factors, which increases the complexity of prediction. In this paper, Multi-Grouped Least Squares Support Vector Machine (MLS-SVM) is proposed for short-term urban traffic flow prediction. In MLS-SVM, spatiotemporal factors (e.g., time, geography, and environment) are divided into different groups. Correlations between each grouped factor are then recognized. Finally, the predicted effect is optimized by combining sub- models for each group. Real-world datasets are used in the experiments of traffic flow prediction. Comparing with the rival methods (i.e., LS-SVM, Wavelet Neural Network, Multi-Factor Pattern Recognition), the simulation results demonstrated the validity and stability of MLS-SVM.

The life of many wireless sensor networks is limited by their battery-based energy source. Recharging batteries from a distance by the wireless energy transferring technique could lift this restriction. However, how to deploy the mobile charging device requires further research. In this paper, we take the energy constraint of mobile wireless charger (MWC) into consideration and aim at minimizing the total energy consumption by it in recharging cycles. After formulating the optimization problem, we present the MMES-LME method based on the modified MAXMIN Ant System and equalization strategy with the constraint of MWC limited energy. The equalization strategy is presented to equalize the lifespan of all sensor nodes to avoid the untimely death of WSN and balance the consumption of MWC travelling energy and recharging energy. Our experimental results demonstrate improved performance in comparison to the greedy method and MM-LME method, which is based on MAX-MIN Ant System.

Sparse representation has been applied to visual tracking. The visual tracking models based on sparse representation use a template set as dictionary atoms to reconstruct candidate samples without considering similarity among atoms. In this paper, we present a robust tracking method based on the inverse sparse group lasso model. Our method exploits both the group structure of similar candidate samples and the local structure between templates and samples. Unlike the conventional sparse representation, the templates are encoded by the candidate samples, and similar samples are selected to reconstruct the template at the group level, which facilitates inter-group sparsity. Every sample group achieves the intra-group sparsity so that the information between the related dictionary atoms is taken into account. Moreover, the local structure between templates and samples is considered to build the reconstruction model, which ensures that the computed coefficients similarity is consistent with the similarity between templates and samples. A gradient descent-based optimization method is employed and a sparse mapping table is obtained using the coefficient matrix and hash-distance weight matrix. Experiments were conducted with publicly available datasets and a comparison study was performed against 20 state-of-the-art methods. Both qualitative and quantitative results are reported. The proposed method demonstrated improved robustness and accuracy and exhibited comparable computational complexity.

Due to the increasing usage of cloud computing applications, it is important to minimize energy cost consumed by a data center, and simultaneously, to improve quality of service via data center management. One promising approach is to switch some servers in a data center to the idle mode for saving energy while to keep a suitable number of servers in the active mode for providing timely service. In this paper, we design both online and offline algorithms for this problem. For the offline algorithm, we formulate data center management as a cost minimization problem by considering energy cost, delay cost (to measure service quality), and switching cost (to change servers’s active/idle mode). Then, we analyze certain properties of an optimal solution which lead to a dynamic programming based algorithm. Moreover, by revising the solution procedure, we successfully eliminate the recursive procedure and achieve an optimal offline algorithm with a polynomial complexity. For the online algorithm, We design it by considering the worst case scenario for future workload. In simulation, we show this online algorithm can always provide near-optimal solutions.

Interference management is an important problem in wireless networks. In this paper, we focus on the successive interference cancellation (SIC) technique, and aim to design an efficient cross-layer solution to increase throughput for multi-hop wireless networks with SIC. We realize that the challenge of this problem is its mixed integer linear programming formulation, which has bunches of integer variables. In order to solve this problem efficiently, we propose an iterative framework to improve the solution for integer variables and use a linear programming to solve the problem for other variables. Our analysis indicates that the proposed algorithm is with polynomial-time complexity. Simulation results show that SIC can increase throughput of a multi-hop wireless network by around 300%.