Efficient Embedding Research Articles

Dynamic Decomposition of Service Function Chain Using a Deep Reinforcement Learning Approach

The Internet of Things (IoT) universe will continue to expand with the advent of the sixth generation of mobile networks (6G), which is expected to support applications and services with higher data rates, ultra-reliability, and lower latency compared to the fifth generation of mobile networks (5G). These new demanding 6G applications will introduce heavy load and strict performance requirements on the network. Network Function Virtualization (NFV) is a promising approach to handling these challenging requirements, but it also poses significant Resource Allocation (RA) challenges. Especially since 6G network services will be highly complicated and comparatively short-lived, network operators will be compelled to deploy these services in a flexible, on-demand, and agile manner. To address the aforementioned issues, microservice approaches are being investigated, in which the services are decomposed and loosely coupled, resulting in increased deployment flexibility and modularity. This study investigates a new RA approach for microservices-based NFV for efficient deployment and decomposition of Virtual Network Function (VNF) onto substrate networks. The decomposition of VNFs involves additional overheads, which have a detrimental impact on network resources; hence, finding the right balance of when and how much decomposition to allow is critical. Thus, we develop a criterion for determining the potential/candidate VNFs for decomposition and also the granularity of such decomposition. The joint problem of decomposition and efficient embedding of microservices is challenging to model and solve using exact mathematical models. Therefore, we implemented a Reinforcement Learning (RL) model using Double Deep Q-Learning, which revealed an almost 50% more normalized Service Acceptance Rate (SAR) for the microservice approach over the monolithic deployment of VNFs.

Tunnel Crack Detection With Linear Seam Based on Mixed Attention and Multiscale Feature Fusion

Crack detection techniques have been rapidly developed in recent years due to the rise of deep learning. However, existing methods struggle to produce accurate crack segmentation results because cracks and linear seams on the tunnel lining surface have significant similarities in terms of intensity value and texture features. At the same time, due to the scarcity of data, the existing tunnel lining surface crack detection methods still use multi-step traditional image processing methods for detection, which is inefficient. In this paper, we collect and label a dataset of 200 tunnel lining surface crack images named Tunnel200. For the first time, a deep learning-based method is used to detect cracks in the tunnel lining surface. To deal with the characteristics of crack and linear seam, which mostly present long strip or curved shapes, we propose a Mixed Attention (MA) module by efficient embedding channel and positional information. Unlike common spatial attention that aggregates information throughout space, mixed attention aggregates feature directly along with two directions, height, and width, in the spatial dimension. In this way, the long-range dependence of the crack features can be effectively captured. The proposed MA is simple to incorporate into the network. Meanwhile, we embed it in the traditional U-shape network while employing an efficient multi-scale feature fusion technique to build the Tunnel Crack Detection Network (TCDNet). TCDNet outperforms other crack detection and semantic segmentation methods on the Tunnel200 dataset. Additionally, we evaluate our method on two publicly available crack datasets, Crack500 and DeepCrack, and our method gets superior performance.

A semi‐supervised network based on feature embeddings for image classification

AbstractDeep learning approaches, including convolutional neural networks, are suitable for image classification tasks with well‐labelled data. Unfortunately, we do not always have sufficiently labelled data. Recent methods attempt to leverage labelled and unlabelled data using fine‐tuning or transfer learning. However, these methods rely on low‐level image features. This article departs from recent works and proposes a new semi‐supervised learning network that constitutes a convolutional branch and a neighbour cluster branch. Also, we introduce a new loss function that carefully optimizes the network according to the labelled/unlabelled data. In this way, we reduce any tendency to rely on low‐level features, which is the case in current methods. We use datasets from three different domains (hand‐written digits, natural images, and objects) to analyse the performance of our method. Experimental analysis shows that the network performs better by learning inherent discrimination features when integrating unlabelled data into the model's training process. Our proposed approach also provides strong generalization in the context of transfer learning. Finally, this study shows that the proposed loss function optimizes the network to produce more efficient feature embeddings for domain adaptation.

Marketing geographical indication products in the digital age: a holistic perspective

PurposeThis paper analyses the impact of digitalization in the marketing of geographic indication (GI) products. Specifically, the objective is to provide a systemic and comprehensive view of marketing issues and challenges arising from evolving digitalization in the agriculture sector.Design/methodology/approachThe authors employed an explorative cognitive mapping technique on a sample of key informants among Italian companies and a consortium of registered food, spirits and wine products.FindingsThis study describes the key concepts dominant in the discourse of informants concerning digitalization and its influence on the GI market. Three clusters and two loops were also identified to explain the relationship among key concepts, which stress changes regarding the relationship with the consumer, product experience and innovation, and the integration between physical and virtual space.Research limitations/implicationsThis is the first attempt to apply cognitive maps to GIs. However, the study does have limitations: it was conducted on a small number of producers and was restricted to only Italy. Another limitation is that interviews were conducted during the COVID-19 pandemic.Practical implicationsThe results suggest important practical implications that stress the need for learning and increased digital competences for the efficient embedding of digital technologies in all business areas, increase product value and innovation, and the need for governance in support of digital transformation.Originality/valueThis work creates the new and valuable literature on food marketing and, specifically, on a changing market environment resulting from digitalization, by providing a holistic overview of digitalization in reference to the marketing of GIs.

Efficient Virtual Network Embedding of Cloud-Based Data Center Networks into Optical Networks

The demand for data center bandwidth has exploded due to the continuous development of cloud computing, causing the use of network resources close to saturation. Optical network has become an encouraging technology for many burgeoning networks and parallel/distributed computing applications because of its huge bandwidth. This article focuses on efficient embedding of data centers into optical networks, which aims to reduce complexity of the network topology by using the parallel transmission characteristics of optical fiber. We first present a novel virtual network embedding (VNE) mathematical model used for optical data center networks. Then we derive a priority of location VNE algorithm according to node proximity sensing and path comprehensive evaluation. Furthermore, we propose routing and wavelength assignment for DCNs into optical networks, and identify the lower bound of the required number of wavelengths. Extensive evaluations show that the proposed embedding algorithm can reduce the average waiting time of virtual network requests by 20 percent, increase the request acceptance rate and revenue-overhead ratio by 13 percent, as compared to the latest VNE algorithm.

Knowledge tensor embedding framework with association enhancement for breast ultrasound diagnosis of limited labeled samples

In the AI diagnosis of breast cancer, instead of ultrasound images from non-standard acquisition process, the Breast Image Reporting and Data System (BI-RADS) reports are widely accepted as the input data since it can give standardized descriptions for the breast ultrasound samples. The BI-RADS reports are usually stored as the format of Knowledge Graph (KG) due to the flexibility, and the KG embedding is a common procedure for the AI analysis on BI-RADS data. However, since most existing embedding methods are based on the local connections in KG, in the situation of limited labeled samples, there is a clear need for embedding based diagnosis method which is capable of representing the global interactions among all entities/relations and associating the labeled/unlabeled samples. To diagnose the breast ultrasound samples with limited labels, in this paper we propose an efficient framework Knowledge Tensor Embedding with Association Enhancement Diagnosis (KTEAED), which adopts tensor decomposition into the embedding to achieve the global representation of KG entities/relations, and introduces the association enhancement strategy to prompt the similarities between embeddings of labeled/unlabeled samples. The embedding vectors are then utilized to diagnose the clinical outcomes of samples by predicting their links to outcomes entities. Through extensive experiments on BI-RADS data with different fractions of labels and ablation studies, our KTEAED displays promising performance in the situations of various fractions of labels. In summary, our framework demonstrates a clear advantage of tackling limited labeled samples of BI-RADS reports in the breast ultrasound diagnosis.

Nonparametric spectral methods for multivariate spatial and spatial–temporal data

We propose computationally efficient methods for estimating stationary multivariate spatial and spatial–temporal spectra from incomplete gridded data. The methods are iterative and rely on successive imputation of data and updating of model estimates. Imputations are done according to a periodic model on an expanded domain. The periodicity of the imputations is a key feature that reduces edge effects in the periodogram and is facilitated by efficient circulant embedding techniques. In addition, we describe efficient methods for decomposing the estimated cross spectral density function into a linear model of coregionalization plus a residual process. The methods are applied to two storm datasets, one of which is from Hurricane Florence, which struck the southeastern United States in September 2018. The application demonstrates how fitted models from different datasets can be compared, and how the methods are computationally feasible on datasets with more than 200,000 total observations.

GCRFLDA: scoring lncRNA-disease associations using graph convolution matrix completion with conditional random field.

Long noncoding RNAs (lncRNAs) play important roles in various biological regulatory processes, and are closely related to the occurrence and development of diseases. Identifying lncRNA-disease associations is valuable for revealing the molecular mechanism of diseases and exploring treatment strategies. Thus, it is necessary to computationally predict lncRNA-disease associations as a complementary method for biological experiments. In this study, we proposed a novel prediction method GCRFLDA based on the graph convolutional matrix completion. GCRFLDA first constructed a graph using the available lncRNA-disease association information. Then, it constructed an encoder consisting of conditional random field and attention mechanism to learn efficient embeddings of nodes, and a decoder layer to score lncRNA-disease associations. In GCRFLDA, the Gaussian interaction profile kernels similarity and cosine similarity were fused as side information of lncRNA and disease nodes. Experimental results on four benchmark datasets show that GCRFLDA is superior to other existing methods. Moreover, we conducted case studies on four diseases and observed that 70 of 80 predicted associated lncRNAs were confirmed by the literature.

Embedding Directed Graphs in Potential Fields Using FastMap-D

Embedding undirected graphs in a Euclidean space has many computational benefits. FastMap is an efficient embedding algorithm that facilitates a geometric interpretation of problems posed on undirected graphs. However, Euclidean distances are inherently symmetric and, thus, Euclidean embeddings cannot be used for directed graphs. In this paper, we present FastMap-D, an efficient generalization of FastMap to directed graphs. FastMap-D embeds vertices using a potential field to capture the asymmetry between the to-and-fro pairwise distances in directed graphs. FastMap-D learns a potential function to define the potential field using a machine learning module. In experiments on various kinds of directed graphs, we demonstrate the advantage of FastMap-D over other approaches.

A QoS-Based Splitting Strategy for a Resource Embedding Across Multiple Cloud Providers

In cloud computing, a fundamental management problem with the Infrastructure as a Service model lies in the efficient embedding of computing and networking resources onto distributed virtualized infrastructures. This issue, usually referred to as the Virtual Network Embedding (VNE) problem, has been well studied for a single Cloud Provider (CP). However, wide-area services delivery may require to embed heterogeneous resources over multiple CPs. This adds more complexity and scalability issues, since the Virtual Network Requests (VNRs) embedding process requires two phases of operation: the multicloud VNRs splitting, followed by the intracloud VNR segments mapping. This paper addresses the problem of VNE across multiple CPs by proposing a VNRs splitting strategy which aims at improving the performance and QoS of resulting VNR segments. An Integer Linear Program (ILP) is used to formalize the splitting phase as a maximization problem with constraints. Subsequently, in order to minimize the overall delay, a multi-objective intracloud resource mapping approach formalized as a Mixed-Integer Linear Program (MILP) is adopted. Simulations with the exact method show the efficiency of the proposed strategy based on several performance criteria. In particular, the acceptance rate and the delay are respectively improved by 15.1 and 18.5 percent, while preventing QoS violations.

Prominence Ranking in Graphs with Community Structure

We consider prominence ranking in graphs involving actors, their artifacts and the artifact groups. When multiple actors contributing to an artifact constitutes a social tie, associations between the artifacts can be used to infer prominence among actors. This is because prominent actors will tend to collaborate on prominent artifacts, and prominent artifacts will be associated with other prominent artifacts. Our testbed example is the DBLP co-authorship graph: multiple authors (the actors) collaborate to publish research papers (the artifacts); collaboration is the social tie. Papers have prominence themselves (eg. quality and impact of the work) and the prominence of the venues are tied to the prominence of the papers in them. We use our methods to infer prominence based on the venue-based associations of papers, and compare our rankings with external citation based measures of prominence. We compare with numerous other ranking algorithms, and show that the ranking performance gain from using the venues is statistically significant. What if there are no natural artifact groups like venues? We develop a new algorithm which uses discovered artifact groups. Our approach consists of two steps. First, we find artifact groups by linking artifacts with common contributors. Note that instead of finding communities of actors, we consider communities of artifacts. We then use these grouped artifacts in the prominence ranking algorithm. We consider different methods for obtaining the artifact groups, in particular a very efficient embedding based algorithm for graph clustering and show the effectiveness of our method in improving the ranking of actors. The inferred groups are as good as or better than the natural conference venues for DBLP.

Efficient and Green Embedding of Virtual Data Centers with Mixture of Unicast and Multicast Services

The improved efficiency achieved by virtualizing data centers (DCs) has been well established. In this paper, we propose a mixed Virtual Data Center (VDC) capable of supporting both unicast and multicast services. We provide a new method to realize the embedding of these VDCs. We also provide a Mixed Integer Linear Programming (MILP) formulation and a scalable heuristic algorithm for efficiently embedding its demands. Numerical results show that mixed VDC embedding supporting both unicast and multicast services performs significantly better than existing embedding methods in terms of system cost, power consumption, link capacity utilization, and VDC acceptance ratio.

An Efficient Approach Based on Ant Colony Optimization and Tabu Search for a Resource Embedding Across Multiple Cloud Providers

In cloud computing, a fundamental management problem with the Infrastructure as a Service (IaaS) model lies in the efficient embedding of computational and networking resources onto distributed virtualized infrastructures owned by independent cloud providers (CPs). In such a context, this issue usually referred to as the Virtual Network Embedding (VNE) problem, adds more complexity since the entire embedding process requires two mayor phases of operation: the multicloud virtual network requests (VNRs) splitting, followed by the intracloud VNR segments mapping. This paper focuses on the splitting phase problem, by proposing a VNRs splitting strategy formalized as an Integer Linear Program (ILP) model, with the objective of improving the performance and QoS of resulting mapped VNR segments, while minimizing the resource provisioning expenditures. As the VNE is classified as an NP-hard problem, a hybrid metaheuristic approach based on the Ant Colony Optimization (ACO) combined with the Tabu Search (TS) as local search operator, is proposed in order to find good feasible solutions in reasonable time. The simulation results show the efficiency of the proposed approach, which generates, in a highly reduced computing time, solution costs very close to the exact solution, with an average cost gap ranging from 0 percent to a maximum of 3.42 percent.

Task Programming: Learning Data Efficient Behavior Representations.

Specialized domain knowledge is often necessary to accurately annotate training sets for in-depth analysis, but can be burdensome and time-consuming to acquire from domain experts. This issue arises prominently in automated behavior analysis, in which agent movements or actions of interest are detected from video tracking data. To reduce annotation effort, we present TREBA: a method to learn annotation-sample efficient trajectory embedding for behavior analysis, based on multi-task self-supervised learning. The tasks in our method can be efficiently engineered by domain experts through a process we call "task programming", which uses programs to explicitly encode structured knowledge from domain experts. Total domain expert effort can be reduced by exchanging data annotation time for the construction of a small number of programmed tasks. We evaluate this trade-off using data from behavioral neuroscience, in which specialized domain knowledge is used to identify behaviors. We present experimental results in three datasets across two domains: mice and fruit flies. Using embeddings from TREBA, we reduce annotation burden by up to a factor of 10 without compromising accuracy compared to state-of-the-art features. Our results thus suggest that task programming and self-supervision can be an effective way to reduce annotation effort for domain experts.

Codee: A Tensor Embedding Scheme for Binary Code Search

Given a target binary function, the binary code search retrieves top-K similar functions in the repository, and similar functions represent that they are compiled from the same source codes. Searching binary code is particularly challenging due to large variations of compiler tool-chains and options and CPU architectures, as well as thousands of binary codes. Furthermore, there are some pivotal issues in current binary code search schemes, including inaccurate text-based or token-based analysis, slow graph matching, or complex deep learning processes. In this paper, we present an unsupervised tensor embedding scheme, Codee, to carry out code search efficiently and accurately at the binary function level. First, we use an NLP-based neural network to generate the semantic-aware token embedding. Second, we propose an efficient basic block embedding generation algorithm based on the network representation learning model. We learn both the semantic information of instructions and the control flow structural information to generate the basic block embedding. Then we use all basic block embeddings in a function to obtain a variable-length function feature vector. Third, we build a tensor to generate function embeddings based on the tensor singular value decomposition, which compresses the variable-length vectors into short fixed-length vectors to facilitate efficient search afterward. We further propose a dynamic tensor compression algorithm to incrementally update the function embedding database. Finally, we use the local sensitive hash method to find the top- <inline-formula><tex-math notation="LaTeX">$K$</tex-math></inline-formula> similar matching functions in the repository. Compared with state-of-the-art cross-optimization-level code search schemes, such as Asm2Vec and DeepBinDiff, our scheme achieves higher average search accuracy, shorter feature vectors, and faster feature generation performance using four datasets, OpenSSL, Coreutils, libgmp and libcurl. Compared with other cross-platform and cross-optimization-level code search schemes, such as Gemini, Safe, the average recall of our method also outperforms others.

Search Efficient Binary Network Embedding

Traditional network embedding primarily focuses on learning a continuous vector representation for each node, preserving network structure and/or node content information, such that off-the-shelf machine learning algorithms can be easily applied to the vector-format node representations for network analysis. However, the learned continuous vector representations are inefficient for large-scale similarity search, which often involves finding nearest neighbors measured by distance or similarity in a continuous vector space. In this article, we propose a search efficient binary network embedding algorithm called BinaryNE to learn a binary code for each node, by simultaneously modeling node context relations and node attribute relations through a three-layer neural network. BinaryNE learns binary node representations using a stochastic gradient descent-based online learning algorithm. The learned binary encoding not only reduces memory usage to represent each node, but also allows fast bit-wise comparisons to support faster node similarity search than using Euclidean or other distance measures. Extensive experiments and comparisons demonstrate that BinaryNE not only delivers more than 25 times faster search speed, but also provides comparable or better search quality than traditional continuous vector based network embedding methods. The binary codes learned by BinaryNE also render competitive performance on node classification and node clustering tasks. The source code of the BinaryNE algorithm is available at https://github.com/daokunzhang/BinaryNE.

An Improved NID Steganography Method for Low Bitrate Speech on VoIP

In the procedure of encoding process on low bitrate speech, fixed codebook division is an efficient and promising embedding method for steganography. An improved neighbor index division (NID) steganography method based on the high bitrate frame of G.723.1 codec (6.3kbit/s) is proposed, which employs the parity and low distortion of neighbor indices for G.723.1 fixed codebooks. Differing from previously NID method which performs quantized index modulation (QIM) beforehand, the proposed method divides codeword indices into separate sub-codebooks according to the secret message bits dynamically in the original G.723.1 codec quantization period. Compared with existing NID method, our proposed method doesn’t need to divide the codebook before the encoding starts. The embedding and codebook dividing happen simultaneously, which utilizes the characteristics of specific secret message bits. The experiment results show that the proposed method has a much lower quality degradation for the decoding speech and still fulfills the low latency requirement for communication.

Learning Pattern Relation-Based Hyperbolic Embedding for Adverse Drug Reaction Extraction

Hyperbolic embedding has been recently developed to allow us to embed words in a Cartesian product of hyperbolic spaces, and its efficiency has been proved in several works of literature since the hierarchical structure is the natural form of texts. Such a hierarchical structure exhibits not only the syntactic structure but also semantic representation. This paper presents an approach to learn meaningful patterns by hyperbolic embedding and then extract adverse drug reactions from electronic medical records. In the experiments, the public source of data from MIMIC-III (Medical Information Mart for Intensive Care III) with over 58,000 observed hospital admissions of the brief hospital course section is used, and the result shows that the approach can construct a set of efficient word embeddings and also retrieve texts of the same relation type with the input. With the Poincaré embeddings model and its vector sum (PC-S), the authors obtain up to 82.3% in the precision at ten, 85.7% in the mean average precision, and 93.6% in the normalized discounted cumulative gain.

Revealing the state space of turbulence using machine learning

We train deep convolutional autoencoders to learn highly efficient embeddings of two-dimensional turbulence. We define a new technique, latent Fourier analysis, that decomposes these representations into a set of interpretable recurrent patterns, and show how these recurrent patterns are closely related to the simple invariant solutions populating the turbulent attractor. By examining a series of bursting episodes with this framework we are able to identify large numbers of new simple invariant solutions that characterize these events and which have avoided previous detection methods.

Secure data transmission in IoTs based on CLoG edge detection

Cyber–physical Systems (CPS) have revolutionized the industry by utilizing the Internet of Things (IoT) for interconnecting various system components. IoT plays a significant role in the development of the smart system paradigm by building a link between physical components with the virtual world for improving smart services and quality of life. It significantly improves people’s activities, be it their personalized health care, living, or the way they monitor control, and organize the businesses. However, data streaming/communication over an open IoT environment creates several security issues that need to be taken care of, for transferring data securely. This paper presents an efficient information embedding solution for ensuring data security in a cyber–physicalnetwork. In this work, a new efficient edge detector called CLoG, (based on Canny and Laplacian of Gaussian detectors) has been developed and is used for the detection of edge areas in digital images. The secret information has been embedded in detected edges. The proposed detector finds finer edge details compared to state-of-art, making it possible to hide more information in a cover image. This, in turn, reduces the number of cover images required to transmit secret data and as such fulfills the requirements of resource-constrained platforms like IoT. Experimental results show that the proposed scheme is capable of providing high-quality stego-images with an average peak signal-to-noise ratio (PSNR) of 48.12 dB for a payload of 2 bits per pixel (bpp) payload. Further, the scheme has been analyzed for its efficacy in terms of histogram analysis, normalized cross-correlation (NCC), and other objective parameters. Besides, we show that the proposed scheme supports blind extraction, has less computational complexity, and is suitable for IoT based systems.