检索结果-内蒙古大学图书馆

SSRN 2024年

作者： Li, Xuli Ma, Yingcang Zhou, Qimin Zhang, Xiaohong School of Science Xi’an Polytechnic University Xi’an710048 China School of Mathematics and Data Science Shaanxi University of Science and Technology Xi’an710021 China

Time series prediction is one of the most important applications of fuzzy cognitive maps. In general, the state of FCMs in forecasting depends only on the state of the previous moment, but in fact it is also affected by the past state. Hence higher-order fuzzy cognitive maps (HFCMs) are proposed based on FCMs considering historical information and have been widely used for time series forecasting. However, using HFCM to deal with sparse and large-scale multivariate time series are still a challenge, while large-scale data makes it difficult to determine the causal relationship between variables because of the increased number of nodes, so it is necessary to explore the relationship between conceptual nodes to guide large-scale HFCMs learning. Therefore, a sparse large-scale HFCMs learning algorithm guided by Spearman correlation coefficient, called SG-HFCM, is proposed in the paper. The SG-HFCM model is specified as follows: First, the solving of HFCMs model is transform into a regression model and an adaptive loss function is utilized to enhance the robustness of the model. Second, l1-norm is used to improve the sparsity of the weight matrix. Third, in order to more accurately characterize the correlation relationship between variables, the Spearman correlation coefficients is add as a regular term to guide the learning of weight matrices. When calculating the Spearman correlation coefficient, through splitting domain interval method, we can better understand the characteristics of the data, and get better correlation in different small intervals, and more accurately characterize the relationship between the variables in order to guide the weight matrix. In addition, in this paper, the ADMM method and quadratic programming method are used to solve the algorithms to get better solutions for the SG-HFCM, where the quadratic programming can well ensure that the range of the weights and obtaining the optimal solution. At last, we compare the proposed SG-HFCM method wit

关键词： Fuzzy Cognitive Maps

来源：评论

学校读者我要写书评

暂无评论

Revisiting primordial magnetic fields through 21-cm physics: Bounds and forecasts

arXiv

引用

arXiv 2024年

作者： Bhaumik, Arko Paul, Debarun Pal, Supratik Physics and Applied Mathematics Unit Indian Statistical Institute 203 B.T. Road Kolkata700108 India Technology Innovation Hub on Data Science Big Data Analytics and Data Curation Indian Statistical Institute 203 B.T. Road Kolkata700108 India

Primordial magnetic fields (PMFs) may significantly influence 21-cm physics via two mechanisms: (i) magnetic heating of the intergalactic medium (IGM) through ambipolar diffusion (AD) and decaying magnetohydrodynamic turbulence (DT), (ii) impact on the star formation rate density (SFRD) through small-scale enhancement of the matter power spectrum. In this analysis, we integrate both of these effects within a unified analytical framework and use it to determine upper bounds on the parameter space of a nearly scale-invariant non-helical PMF in the light of the global 21-cm signal observed by EDGES. Our findings reveal that the joint consideration of both effects furnishes constraints of the order B0 O(10−2) nG on the present-day magnetic field strength, which are considerably tighter compared to earlier analyses. We subsequently explore the prospects of detecting such a magnetized 21-cm power spectrum at the upcoming SKA-Low mission. For the relevant parameters of the PMF (B0 and nB) and the excess radio background (ξ), SNR estimation and Fisher forecast analysis indicate that it may be possible to constrain these three parameters with relative 1σ uncertainties 10% and an associated SNR 10 at SKA-Low. This also leads to possible correlations among these three parameters, thus revealing intriguing trends of interplay among the various physical processes involved. Copyright © 2024, The Authors. All rights reserved.

关键词： Magnetoplasma

来源：评论

学校读者我要写书评

暂无评论

An Image Super Resolution Algorithm Using Reparameterization 18th

An Image Super Resolution Algorithm Using Reparameterization

引用

18th Chinese Intelligent Systems Conference, CISC 2022

作者： Cai, Qiang Chang, Haodong Li, Haisheng Wan, Ruyi Chen, Xiaoming School of Computer Science and Engineering Beijing Technology and Business University Beijing100048 China Beijing Key Laboratory of Big Data Technology for Food Safety Beijing100048 China National Engineering Laboratory for Agri-Product Quality Traceability Beijing100048 China

ISBN: (纸本)9789811962257

At present, the residual-based deep network model mainly adopts a multi-branch structure, but this structure increases the memory access cost, and the reasoning time is not as good as that of a single-branch network. In this work, we propose a re-parameterized construction structure based on EDSR, namely Re-parameterized Deep Super-Resolution (RepEDSR). The training and inference processes are decoupled to speed up inference without sacrificing performance. Extensive quantitative and qualitative evaluations on benchmark datasets show that the proposed algorithm outperforms existing methods in terms of speed and accuracy. © 2022, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

关键词： Optical resolving power

来源：评论

学校读者我要写书评

暂无评论

Sensitivity Analysis of CoVaR

Sensitivity Analysis of CoVaR

引用

IEEE International Conference on Automation science and Engineering (CASE)

作者： Han Fu L. Jeff Hong Guangxin Jiang School of Data Science Fudan University Shanghai China School of Management and School of Data Science Fudan University Shanghai China School of Management Harbin Institute of Technology Harbin Heilongjiang China

CoVaR is an important measure for assessing the systemic risk of a network composed of many systems. To optimize and control the systemic risk of the network, we need to know the sensitivity of CoVaR. In this paper, we derive closed-form expressions of the CoVaR sensitivities and design batched estimators using the infinitesimal perturbation analysis (IPA) and finite-difference methods. We establish the consistency and asymptotic normality of the proposed estimators and show that the convergence rate of the estimators is strictly slower than $n^{-1/6}$ . Numerical experiments show the effectiveness of our estimator and support the theoretical results.

关键词：

来源：评论

学校读者我要写书评

暂无评论

引用

2023 International Conference on Ubiquitous Communication, Ucom 2023

作者： Lin, Dongting Liu, Yuan Yang, Jiarong Zheng, Beixiong Chen, Fangjiong Chen, Wen Zhang, Rui School of Electronic and Information Engineering South China University of Technology Guangzhou China Shanghai Jiao Tong University Department of Electronics Engineering Shanghai China The Chinese University of Hong Kong Shenzhen China Shenzhen Research Institute of Big Data Shenzhen China

ISBN: (纸本)9798350340433

It is commonly believed that 6G will be established on ubiquitous artificial intelligence (AI). Federated learning is recognized as one of the vital solutions for 6G intelligent applications due to its distributed AI structure and privacy preservation nature. In this paper, we consider an integrated federated and centralized learning network that consists of multiple users and a server for acquiring perceived data from mobile devices. The data similarity rule is adopted at users, where the received data offloaded from devices is either retained at users for distributed training or relayed to server for centralized training, and thus the server also aggregates local models from users. By optimizing bandwidth resource allocation and the number of communication rounds, we formulate an optimization problem with the goal of energy consumption minimization under the constraints of model loss and communication delay. Simulation results show that the proposed scheme not only improves model performance but also reduces energy consumption. © 2023 IEEE.

关键词： Resource allocation

来源：评论

学校读者我要写书评

暂无评论

A Novel Adaptive Perturbation-Based Method for Long-Tailed Open Sets Recognition

SSRN

引用

SSRN 2023年

作者： Ngueilbaye, Alladoumbaye Zhang, Qin Yang, Hongye Zhang, Xiaolin Zhang, Peng Chen, Xiaojun Huang, Joshua Zhexue Big Data Institute College of Computer Science and Software Engineering Shenzhen University Shenzhen518060 China National Engineering Laboratory for Big Data System Computing Technology Shenzhen University Shenzhen518060 China College of Computer Science and Engineering Shandong University of Science and Technology Qingdao266590 China

Real-world data often have a long-tailed and open-ended distribution. A reliable practical machine learning system need to learn from the majority classes and also generalize to minority *** achieve this, acknowledge the minority and the novel class samples, and avoid misclassifying them into one of the majority classes are of great significance. Long-tailed Open Set Recognition aims to learn from the naturally distributed long-tailed and open-ended data, by using the easy-gained head class samples to learn the fundamental information, and also try to improve the out-of-distribution generalization ability by identifying the tail class and open-set class samples as novelty to avoid the misclassification. The key challenges for long-tailed open set recognition is the tail and open-set sensitivity, since they can be easily biased towards dominant classes and perform ineffectively on close-set dataset classes. Inspired by techniques on open set recognition, in this paper, we propose a Class-aware Input Preprocessing (CLAIP) approach to improve the detection of the Out-of-distribution samples by perturbing the input images. In this approach, we employ two techniques to leverage its performance. First, we introduce the middle class samples as auxiliary data during the network training process to force each category's boundaries. Second, we add a Class-aware Input Preprocessing module that can be easily plugged into the testing process by perturbing the salient regions of the input images to further improve the out-of-distribution generalization. The experimental results show that CLAIP achieves high quality and significant performance compared to the existing solutions. © 2023, The Authors. All rights reserved.

关键词： Image enhancement

来源：评论

学校读者我要写书评

暂无评论

Stabilize the latent space for image autoregressive modeling: a unified perspective 24

Stabilize the latent space for image autoregressive modeling...

引用

Proceedings of the 38th International Conference on Neural Information Processing Systems

作者： Yongxin Zhu Bocheng Li Hang Zhang Xin Li Linli Xu Lidong Bing School of Data Science University of Science and Technology of China and State Key Laboratory of Cognitive Intelligence School of Computer Science and Technology University of Science and Technology of China and State Key Laboratory of Cognitive Intelligence Zhejiang University

ISBN: (纸本)9798331314385

Latent-based image generative models, such as Latent Diffusion Models (LDMs) and Mask Image Models (MIMs), have achieved notable success in image generation tasks. These models typically leverage reconstructive autoencoders like VQGAN or VAE to encode pixels into a more compact latent space and learn the data distribution in the latent space instead of directly from pixels. However, this practice raises a pertinent question: Is it truly the optimal choice? In response, we begin with an intriguing observation: despite sharing the same latent space, autoregressive models significantly lag behind LDMs and MIMs in image generation. This finding contrasts sharply with the field of NLP, where the autoregressive model GPT has established a commanding presence. To address this discrepancy, we introduce a unified perspective on the relationship between latent space and generative models, emphasizing the stability of latent space in image generative modeling. Furthermore, we propose a simple but effective discrete image tokenizer to stabilize the latent space for image generative modeling by applying K-Means on the latent features of self-supervised learning models. Experimental results show that image autoregressive modeling with our tokenizer (DiGIT) benefits both image understanding and image generation with the next token prediction principle, which is inherently straightforward for GPT models but challenging for other generative models. Remarkably, for the first time, a GPT-style autoregressive model for images outperforms LDMs, which also exhibits substantial improvement akin to GPT when scaling up model size. Our findings underscore the potential of an optimized latent space and the integration of discrete tokenization in advancing the capabilities of image generative models. The code is available at https://***/DAMO-NLP-SG/DiGIT.

关键词：

来源：评论

学校读者我要写书评

暂无评论

A Multiresolution Method for Non-Contact Heart Rate Estimation Using Facial Video Frames 7

A Multiresolution Method for Non-Contact Heart Rate Estimati...

引用

7th International Conference on Wireless Communications, Signal Processing and Networking, WiSPNET 2022

作者： Das, Mousumi Choudhary, Tilendra Bhuyan, M.K. Sharma, L.N. Pallab Jyoti Dutta, H. Mehta Family School of Data Science and Artificial Intelligence Indian Institute of Technology Department of Electronics and Electrical Engineering Assam Guwahati India

ISBN: (纸本)9781665496483

In recent years, camera-based non-contact heart rate (HR) measurement technology has grown immensely. The system captures the reflection of light from the facial tissues and lead to the formation of a remote photoplethysmogram (rPPG) signal that can be used to measure physiological parameters for cardiac health assessment. Due to environmental interferences, extraction of a reliable rPPG signal is a challenging task and thus, requires a robust denoising algorithm. In this paper, a discrete wavelet transform (DWT)-based multiresolution method is used to remove the noises from the video frames caused due to illumination variation and motion artifacts. Subsequently, rPPG signal is extracted and HR is measured from two region of interests (ROIs), facial and forehead regions. The study evaluates the performance of the proposed method on each of the RGB color channels from both the ROIs. The performance results for the COHFACE dataset show that the proposed method works well for the estimation of HR values. Furthermore, they reveal that the forehead region on the green channel is more suitable for HR measurement. © 2022 IEEE.

关键词： Discrete wavelet transforms

来源：评论

学校读者我要写书评

暂无评论

Texture, Shape and Order Matter: A New Transformer Design for Sequential DeepFake Detection

Texture, Shape and Order Matter: A New Transformer Design fo...

引用

IEEE Workshop on Applications of Computer Vision (WACV)

作者： Yunfei Li Yuezun Li Xin Wang Baoyuan Wu Jiaran Zhou Junyu Dong School of Computer Science and Technology Ocean University of China University at Albany SUNY School of Data Science The Chinese University of Hong Kong Shenzhen China

ISBN: (数字)9798331510831

ISBN: (纸本)9798331510848

Sequential DeepFake detection is an emerging task that predicts the manipulation sequence in order. Existing methods typically formulate it as an image-to-sequence problem, employing conventional Transformer architectures. However, these methods lack dedicated design and consequently result in limited performance. As such, this paper describes a new Transformer design, called TSOM, by exploring three perspectives: Texture, Shape, and Order of Manipulations. Our method features four major improvements: we describe a new texture-aware branch that effectively captures subtle manipulation traces with a Diversiform Pixel Difference Attention module. Then we introduce a Multi-source Cross-attention module to seek deep correlations among spatial and sequential features, enabling effective modeling of complex manipulation traces. To further enhance the cross-attention, we describe a Shape-guided Gaussian mapping strategy, providing initial priors of the manipulation shape. Finally, observing that the subsequent manipulation in a sequence may influence traces left in the preceding one, we intriguingly invert the prediction order from forward to backward, leading to notable gains as expected. Extensive experimental results demonstrate that our method outperforms others by a large margin, highlighting the superiority of our method.

关键词： Deepfakes Computer vision Correlation Shape Annotations Design methodology Transformers

来源：评论

学校读者我要写书评

暂无评论

It Takes Two to Tango: Serverless Workflow Serving via Bilaterally Engaged Resource Adaptation

arXiv

引用

arXiv 2025年

作者： Wu, Jing Wang, Lin Deng, Quanfeng Yu, Chen Zhang, Dong Yan, Bingheng Liu, Fangming National Engineering Research Center for Big Data Technology and System Services Computing Technology and System Lab Cluster and Grid Computing Lab Huazhong University of Science and Technology China Paderborn University Germany Inspur Data Co. Ltd. China Peng Cheng Laboratory China

Serverless platforms typically adopt an early-binding approach for function sizing, requiring developers to specify an immutable size for each function within a workflow beforehand. Accounting for potential runtime variability, developers must size functions for worst-case scenarios to ensure service-level objectives (SLOs), resulting in significant resource inefficiency. To address this issue, we propose Janus, a novel resource adaptation framework for serverless platforms. Janus employs a late-binding approach, allowing function sizes to be dynamically adapted based on runtime conditions. The main challenge lies in the information barrier between the developer and the provider: developers lack access to runtime information, while providers lack domain knowledge about the workflow. To bridge this gap, Janus allows developers to provide hints containing rules and options for resource adaptation. Providers then follow these hints to dynamically adjust resource allocation at runtime based on real-time function execution information, ensuring compliance with SLOs. We implement Janus and conduct extensive experiments with real-world serverless workflows. Our results demonstrate that Janus enhances resource efficiency by up to 34.7% compared to the state-of-the-art. © 2025, CC BY-NC-ND.

关键词： Resource allocation

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：