Rethinking Bayesian Learning for Data Analysis: The art of prior and inference in sparsity-aware modeling

Cheng, Lei; Yin, Feng; Theodoridis, Sergios; Chatzis, Sotirios; Chang, Tsung-Hui

doi:10.1109/msp.2022.3198201

Cited by 78 publications

(30 citation statements)

References 54 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Here, virtual nodes are embedded in physical nodes and virtual links between physical paths by considering some constraints. One of the most important issues in VNE is SFC placement for resource allocation, which is modeled as NP‐hard 13,38 . In general, the embedding process in VNE includes two steps: the node embedding and the link embedding.…”

Section: Proposed Algorithm: Distributed Parallel Chainingmentioning

confidence: 99%

“…One of the most important issues in VNE is SFC placement for resource allocation, which is modeled as NP-hard. 13,38 In general, the embedding process in VNE includes two steps: the node embedding and the link embedding. In the first step, each virtual node must be mapped to an underlying node to provide request resources.…”

Section: Proposed Algorithm: Distributed Parallel Chainingmentioning

confidence: 99%

See 1 more Smart Citation

Dynamic service function chains placement based on parallelized requests in edge computing environment

Guo,

Rezaeipanah

2023

Trans Emerging Tel Tech

View full text Add to dashboard Cite

The advent of Network Function Virtualization (NFV) technology brings flexible traffic engineering to edge computing environments. Online services in NFV are chained as Service Function Chains (SFCs), which consist of ordered sequences of Virtual Network Functions (VNFs). The SFC Placement (SFCP) problem is solved under Quality of Service (QoS) requirements and limited resource availability by directing traffic to the required VNFs. However, SFC assembly leads to high latency and network congestion by increasing the count of VNFs, which parallelized SFC can overcome this problem. With parallelizing an SFC request, independent VNFs are activated simultaneously and computational acceleration is realized by reducing the SFC length. Any pair of VNFs that do not conflict with traffic can be activated simultaneously. Most VNFs are deployed on distributed servers for load balancing, which makes SFC parallelization challenging. Meanwhile, the cost of merging/duplicating packets for parallelized SFCs between different servers is not negligible. Hence, in this article, Distributed Parallel Chaining (DPC) is proposed which is an algorithm based on Deep Reinforcement Learning (DRL) approaches. The DPC algorithm solves the SFCP problem to maximize the Long‐Term Expected Cumulative Reward (LTECR). DPC incorporates an Asynchronous Advantage Actor‐Critic (A3C) algorithm as a new approach based on DRL to increase the admission ability of future SFC requests by maximizing LTECR. The evaluation results show the effectiveness of the proposed algorithm from different aspects. Specifically, compared to the best existing approaches, DPC can reduce SFC latency by 8.7%.

show abstract

Section: Proposed Algorithm: Distributed Parallel Chainingmentioning

confidence: 99%

Section: Proposed Algorithm: Distributed Parallel Chainingmentioning

confidence: 99%

Dynamic service function chains placement based on parallelized requests in edge computing environment

Guo,

Rezaeipanah

2023

Trans Emerging Tel Tech

View full text Add to dashboard Cite

show abstract

“…For instance, Bayesian learning typically resorts to maximizing the logarithm of the p(Y|θ) w.r.t. the model parameters θ, which can automatically trade off between data fit and model complexity, in line with Occam's razor rule [55]. However, p(Y|θ) is obtained by integrating out all the latent variables {X, F, Ũ} in the joint distribution (c.f.…”

Section: A Variational Inference and Approximationsmentioning

confidence: 99%

“…This intractability issue is addressed in the variational Bayesian methods [20] by assuming a variational distribution, q(X, F, Ũ), to approximate the intractable p(X, F, Ũ|Y). With the variational distribution, q(X, F, Ũ), it can be shown that the logarithm of the evidence function is lower bounded by the so-called evidence lower bound (ELBO) [55],…”

Section: A Variational Inference and Approximationsmentioning

confidence: 99%

Towards Flexibility and Interpretability of Gaussian Process State-Space Model

Lin¹,

Yin²

2023

Preprint

View full text Add to dashboard Cite

Gaussian process state-space model (GPSSM) has attracted much attention over the past decade. However, the model representation power of GPSSM is far from satisfactory. Most GPSSM works rely on the standard Gaussian process (GP) with a preliminary kernel, such as squared exponential (SE) kernel and Matérn kernel, which limit the model representation power and its application in complex scenarios. To address this issue, this paper proposes a novel class of probabilistic statespace model named TGPSSM that enriches the GP priors in the standard GPSSM through parametric normalizing flow, making the state-space model more flexible and expressive. In addition, by inheriting the advantages of sparse representation of GP models, we propose a scalable and interpretable variational learning algorithm to learn the TGPSSM and infer the latent dynamics simultaneously. By integrating a constrained optimization framework and explicitly constructing a non-Gaussian state variational distribution, the proposed learning algorithm enables the TG-PSSM to significantly improve the capabilities of state space representation and model inference. Experimental results based on various synthetic and real datasets corroborate that the proposed TGPSSM yields superior learning and inference performance compared to several state-of-the-art methods. The accompanying source code is available at https://github.com/zhidilin/TGPSSM.

show abstract

“…The technology of faking celebrities’ essential pictures and videos makes society’s most significant issue. Pornographic pictures and images of politicians are modified to spread a wrong impression about the targeted person in the community ( Rössler et al, 2019 ; Zhou et al, 2019 ; Cheng et al, 2022 ). Sometimes fraudsters misuse the images with sensitive issues.…”

Section: Introductionmentioning

confidence: 99%

VIOLA jones algorithm with capsule graph network for deepfake detection

Venkatachalam

Trojovský

Hubálovský

2023

PeerJ Computer Science

View full text Add to dashboard Cite

DeepFake is a forged image or video created using deep learning techniques. The present fake content of the detection technique can detect trivial images such as barefaced fake faces. Moreover, the capability of current methods to detect fake faces is minimal. Many recent types of research have made the fake detection algorithm from rule-based to machine-learning models. However, the emergence of deep learning technology with intelligent improvement motivates this specified research to use deep learning techniques. Thus, it is proposed to have VIOLA Jones’s (VJ) algorithm for selecting the best features with Capsule Graph Neural Network (CN). The graph neural network is improved by capsule-based node feature extraction to improve the results of the graph neural network. The experiment is evaluated with CelebDF-FaceForencics++ (c23) datasets, which combines FaceForencies++ (c23) and Celeb-DF. In the end, it is proved that the accuracy of the proposed model has achieved 94.

show abstract

Rethinking Bayesian Learning for Data Analysis: The art of prior and inference in sparsity-aware modeling

Cited by 78 publications

References 54 publications

Dynamic service function chains placement based on parallelized requests in edge computing environment

Dynamic service function chains placement based on parallelized requests in edge computing environment

Towards Flexibility and Interpretability of Gaussian Process State-Space Model

VIOLA jones algorithm with capsule graph network for deepfake detection

Contact Info

Product

Resources

About