检索结果-内蒙古大学图书馆

2025 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2025

作者： Dong, Jinwei Wang, Xinsheng Mao, Qirong School of Computer Science and Communication Engineering Jiangsu University China Jiangsu Engineering Research Center of Big Data Ubiquitous Perception and Intelligent Agriculture Applications China Provincial Key Laboratory of Computational Intelligence and New Technologies in Low-Altitude Digital Agriculture Zhenjiang China Audio Speech and Language Processing Group School of Computer Science Northwestern Polytechnical University Xi'an China

ISBN: (纸本)9798350368741

Generative models have attracted considerable attention for speech separation tasks, and among these, diffusion-based methods are being explored. Despite the notable success of diffusion techniques in generation tasks, their adaptation to speech separation has encountered challenges, notably slow convergence and suboptimal separation outcomes. To address these issues and enhance the efficacy of diffusion-based speech separation, we introduce EDSep, a novel single-channel method grounded in score matching via stochastic differential equation (SDE). This method enhances generative modeling for speech source separation by optimizing training and sampling efficiency. Specifically, a novel denoiser function is proposed to approximate data distributions, which obtains ideal denoiser outputs. Additionally, a stochastic sampler is carefully designed to resolve the reverse SDE during the sampling process, gradually separating speech from mixtures. Extensive experiments on databases such as WSJ0-2mix, LRS2-2mix, and VoxCeleb2-2mix demonstrate our proposed method's superior performance over existing diffusion and discriminative models, validating its efficacy. © 2025 IEEE.

关键词： diffusion score matching speech separation stochastic differential equation

来源：评论

学校读者我要写书评

暂无评论

A generalized complex power iteration-based algorithm for quartic–quadratic optimization problems over a spherical constraint and its applications

引用

Journal of Computational and Applied Mathematics 2025年 470卷

作者： Tang, Jie Sheng, Zhou Li, Xin Department of Data Science Anhui University of Technology Maanshan Anhui243032 China Anhui Provincial Joint Key Laboratory of Disciplines for Industrial Big Data Analysis and Intelligent Decision Maanshan Anhui243032 China School of Mathematics Hefei University of Technology Hefei Anhui230009 China

Quartic–quadratic optimization problems over the sphere have received a great attention in various real problems. Observations of three concrete applications primarily trigger our interest in this problem. This paper considers a generalized complex power method to solve the problem. Such a nonconvex optimization problem is studied by giving a suitable shift term to ensure convexity. Several convergence results are established, and we analyze the worst-case complexity in terms of the step-size and the number of iterations. We then apply our method to three tasks: (1) computing the ground state of Bose–Einstein condensation (BEC);(2) computing the quartic–quadratic optimization problems over the ellipsoid, which arises in the problem of computing the ground state of BEC by using a new discretization;and (3) approximating the desired transmit beampattern. The reported numerical results indicate that our method performs relatively quickly and effectively. © 2025 Elsevier B.V.

关键词： Iterative methods

来源：评论

学校读者我要写书评

暂无评论

Lyapunov-guided Deep Reinforcement Learning for Semantic-aware AoI Minimization in UAV-assisted Wireless Networks

引用

IEEE Transactions on Wireless Communications 2025年

作者： Long, Yusi Gong, Shimin Sun, Sumei Lee, Gary C.F Li, Lanhua Niyato, Dusit Shenzhen Campus of Sun Yat-sen University School of Intelligent Systems Engineering Shenzhen518000 China Guangdong Provincial Key Laboratory of Fire Science and Intelligent Emergency Technology China A*STAR Singapore Nanyang Technological University College of Computing and Data Science Singapore

This paper investigates an unmanned aerial vehicle (UAV)-assisted semantic network where the ground users (GUs) periodically capture and upload the sensing information to a base station (BS) via UAVs’ relaying. Both the GUs and the UAVs can extract semantic information from large-size raw data and transmit it to the BS for recovery. Smaller-size semantic information reduces latency and improves information freshness, while larger-size semantic information enables more accurate data reconstruction at the BS, preserving the value of original information. We introduce a novel semantic-aware age-of-information (SAoI) metric to capture both information freshness and semantic importance, and then formulate a time-averaged SAoI minimization problem by jointly optimizing the UAV-GU association, the semantic extraction, and the UAVs’ trajectories. We decouple the original problem into a series of subproblems via the Lyapunov framework and then use hierarchical deep reinforcement learning (DRL) to solve each subproblem. Specifically, the UAV-GU association is determined by DRL, followed by the optimization module updating the semantic extraction strategy and UAVs’ deployment. Simulation results show that the hierarchical structure improves learning efficiency. Moreover, it achieves low AoI through semantic extraction while ensuring minimal loss of original information, outperforming the existing baselines. © 2002-2012 IEEE.

关键词： Deep reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Channel and space-based joint rate allocation algorithm

Channel and space-based joint rate allocation algorithm

引用

International Conference on Acoustics, Speech, and Signal Processing (ICASSP)

作者： Dayong Wang Chao Yuan Yu Sun Xin Lu Hui Guo Frederic Dufaux Ce Zhu Key Laboratory of Big Data Intelligent Computing Chongqing University of Posts and Telecommunications Guangxi Key Laboratory of Machine Vision and Intelligent Control Wuzhou University Chongqing Key Laboratory of Image Cognition Chongqing University of Posts and Telecommunications China Department of Computer Science University of Central Arkansas Faculty of Computing Engineering and Media (CEM) De Montfort University UK Université Paris-Saclay CNRS CentraleSupélec Laboratoire Des Signaux et Systèmes France School of Information and Communication Engineering University of Electronic Science and Technology of China

ISBN: (数字)9798350368741

ISBN: (纸本)9798350368758

Rate control is a critical component for image and video compression Particularly under limited network bandwidth conditions, bitrate control is essential to ensure efficient image transmission by effectively allocation channel resources. In this research, since both Channel and Spatial have relationship with rate allocation, we first propose a joint Channel-wise and Spatial-wise Quantization scheme to determine optimal quantization parameters. Subsequently, we develop a quantization step estimation network to obtain parameters to efficiently allocate rate according to target rate. Experiments demonstrate that our algorithm significantly improve compressed image quality with minimal bitrate distortion and achieve accurate rate control with nearly 3% average bitrate error.

关键词： Image quality Quantization (signal) Image coding Bit rate Signal processing algorithms Estimation Rate-distortion Video compression Resource management Speech processing

来源：评论

学校读者我要写书评

暂无评论

LLM Knows Geometry Better than Algebra: Numerical Understanding of LLM-Based Agents in A Trading Arena

arXiv

引用

arXiv 2025年

作者： Ma, Tianmi Du, Jiawei Huang, Wenxin Wang, Wenjie Xie, Liang Zhong, Xian Zhou, Joey Tianyi Hubei Key Laboratory of Transportation Internet of Things Wuhan University of Technology China Hubei Key Laboratory of Big Data Intelligent Analysis and Application Hubei University China Centre for Frontier AI Research Agency for Science Technology and Research Singapore Institute of High Performance Computing Agency for Science Technology and Research Singapore School of Computing National University of Singapore Singapore School of Science Wuhan University of Technology China

Recent advancements in large language models (LLMs) have significantly improved performance in natural language processing tasks. However, their ability to generalize to dynamic, unseen tasks, particularly in numerical reasoning, remains a challenge. Existing benchmarks mainly evaluate LLMs on problems with predefined optimal solutions, which may not align with real-world scenarios where clear answers are absent. To bridge this gap, we design the Agent Trading Arena, a virtual numerical game simulating complex economic systems through zero-sum games, where agents invest in stock portfolios. Our experiments reveal that LLMs, including GPT-4o, struggle with algebraic reasoning when dealing with plain-text stock data, often focusing on local details rather than global trends. In contrast, LLMs perform significantly better with geometric reasoning when presented with visual data, such as scatter plots or K-line charts, suggesting that visual representations enhance numerical reasoning. This capability is further improved by incorporating the reflection module, which aids in the analysis and interpretation of complex data. We validate our findings on NASDAQ STOCK dataset, where LLMs demonstrate stronger reasoning with visual data compared to text. Our code and data are publicly available at https://***/wekjsdvnm/***. © 2025, CC BY.

关键词： Natural language processing systems

来源：评论

学校读者我要写书评

暂无评论

Collision-Aware MPCC Integrated with Neural Augmentation for Quadrotor Trajectory Tracking

Collision-Aware MPCC Integrated with Neural Augmentation for...

引用

2025 IEEE International Conference on Mechatronics, ICM 2025

作者： Hu, Chuanqi Wang, Jin Chen, Yikai Chen, Yonghang Zheng, Zhi Meng, Jun Lu, Guodong Zhejiang University State Key Laboratory of Fluid Power and Mechatronic Systems Hangzhou310058 China Zhejiang University Zhejiang Key Laboratory of Industrial Big Data and Robot Intelligent Systems Hangzhou310058 China Robotics Research Center of Yuyao City Ningbo315400 China National Elite Institute of Engineering Chongqing University Chongqing401135 China School of Mechanical Engineering Zhejiang University Hangzhou310058 China College of Electrical Engineering Zhejiang University Hangzhou310027 China Center for Data Mining and Systems Biology College of Electrical Engineering Zhejiang University Hangzhou310027 China

ISBN: (纸本)9798331533892

This paper presents a trajectory tracking algorithm designed for autonomous quadrotors to navigate complex dy-namic environments that is resilient to collision disturbances. Since it combines Model Predictive Contouring Control (MPCC) with the capabilities of neural networks, our method is named MPCC Integrated with Neural Augmentation (MINA). By equip-ping the quadrotor with PVDF piezoelectric sensors, we achieve real-time perception of collision forces and utilize S4D-Conv Adaptive Net Architecture (SANA) for state computation, a network that integrates the advantages of S4D and convolutional kernels to enhance the accuracy of state predictions. We incorporate the predicted results into the MPCC framework to optimize the thrust output of the quadrotor for the next time step, ensuring precise trajectory tracking. Experimental validation demonstrates the reliability and effectiveness of our algorithm, providing an innovative solution for autonomous flight and trajectory tracking of quadrotors in complex dynamic environments. © 2025 IEEE.

关键词： Flight paths

来源：评论

学校读者我要写书评

暂无评论

Multi-feature Based Memory-Enhanced Knowledge Tracing 6th

Multi-feature Based Memory-Enhanced Knowledge Tracing

引用

6th International Conference on Computer Science and Educational Informatization, CSEI 2024

作者： Sun, Youwei Liu, Ruyi Miao, Qiguang Lu, Zixiang Zhao, Peipei Li, Ronghan School of Computer Science and Technology Xidian University Shaanxi Xi’an710071 China Xi’an Key Laboratory of Big Data and Intelligent Vision Shaanxi Xi’an710071 China Key Laboratory of Collaborative Intelligence Systems Ministry of Education Xidian University Xi’an710071 China

ISBN: (纸本)9789819637379

In the realm of intelligent education, knowledge tracing (KT) has emerged as a crucial tool for enhancing educational efficiency and enabling personalized learning. Traditional KT methods primarily rely on students’ historical response data to evaluate their mastery of specific skills and forecast future performance. However, focusing solely on basic interactions, skills and response outcomes, makes it challenging to capture the complexities of real learning environments and accurately reflect students’ genuine knowledge states. To address this issue, we propose an innovative multi-feature based memory-enhanced knowledge tracing (MMKT) method. The key innovation of the MMKT model lies in the introduction of ‘memory features’ and ‘interaction features’. Memory features account for features such as the interval between repeated responses, the total number of responses, and memory interference during the overall learning process. In contrast, interaction features include response time, number of attempts, and hints. To integrate these features, we employ a self-attention mechanism coupled with a memory enhancement module and a recurrent neural network. This approach effectively uncovers variations in knowledge states while balancing the influence of global and local interactions. Our experimental results from different public datasets emphasize the benefits of the MMKT model in predicting future performance, underscoring the significance of integrating multi-feature and real-world relevant memory improvements. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Students

来源：评论

学校读者我要写书评

暂无评论

Upconversion Luminescence and Optical Thermometry Behaviors of Yb3+ and Ho3+ Co-Doped Nabi(Moo4)2 Single Crystal

SSRN

引用

SSRN 2025年

作者： Su, Wenzhi Zhang, Mengyu Zhang, Chuancheng Zou, Yong Liu, Wenpeng Ding, Shoujun School of Microelectronics and Data Science Anhui University of Technology Anhui Maanshan243002 China Anhui Provincial Joint Key Laboratory of Disciplines for Industrial Big Data Analysis and Intelligent Decision Anhui Maanshan243002 China Anhui Hefei230031 China

Upconversion (UC) fluorescence intensity ratio (FIR) technology has emerged as a promising approach for optical thermometry due to its ability to provide non-contact, high-precision temperature measurements. Unlike conventional UC phosphor-based systems, single-crystal UC materials offer enhanced responsiveness and sensitivity for thermal detection, thanks to their structural stability and superior thermal conductivity. In this study, we successfully synthesized an Yb3+, Ho3+:NaBi(MoO4)2 (NBM) single crystal with exceptional optical quality using the Czochralski (Cz) growth technique. Comprehensive structural characterization was carried out using X-ray diffraction (XRD), Fourier transform infrared (FTIR) spectroscopy, and scanning electron microscopy (SEM). The results confirmed that Yb3+, Ho3+:NBM crystallizes in a tetragonal structure with the space group I41/a. The emission of different visible light colors under various excitation wavelengths was found to be due to the multiphonon relaxation process. Systematic analysis of power-dependent up-conversion emission spectra conclusively showed that the green and red up-conversion emissions arise from two-photon excitation processes. Moreover, temperature-dependent spectral evolution in the range of 300-500 K was quantitatively analyzed, allowing the calculation of both absolute sensitivity (Sa) and relative sensitivity (Sᵣ) parameters to evaluate the material's optical thermometric performance. These findings highlight the promising potential of Yb3+, Ho3+:NBM single crystal for developing high-performance optical thermometry devices. © 2025, The Authors. All rights reserved.

关键词： Single crystals

来源：评论

学校读者我要写书评

暂无评论

SAV-SE: Scene-aware Audio-Visual Speech Enhancement with Selective State Space Model

引用

IEEE Journal on Selected Topics in Signal Processing 2025年

作者： Qian, Xinyuan Gao, Jiaran Zhang, Yaodan Zhang, Qiquan Liu, Hexin Garcia, Leibny Paola Li, Haizhou University of Science and Technology Beijing School of Computer and Communication Engineering Beijing100083 China University of New South Wales School of Electrical Engineering and Telecommunications Sydney2052 Australia Nanyang Technological University College of Computing and Data Science Singapore Singapore Johns Hopkins University CLSP and HLT-COE United States Chinese University of HongKong Guangdong Provincial Key Laboratory of Big Data Computing Shenzhen518172 China Shenzhen Research Institute of Big data Shenzhen51872 China

Speech enhancement plays an essential role in various applications, and the integration of visual information has been demonstrated to bring substantial advantages. However, the majority of current research concentrates on the examination of facial and lip movements, which can be compromised or entirely inaccessible in scenarios where occlusions occur or when the camera view is distant. Whereas contextual visual cues from the surrounding environment have been overlooked: for example, when we see a dog bark, our brain has the innate ability to discern and filter out the barking noise. To this end, in this paper, we introduce a novel task, i.e. Scene-aware Audio-Visual Speech Enhancement SAV-SE. To our best knowledge, this is the first proposal to use rich contextual information from synchronized video as auxiliary cues to indicate the type of noise, which eventually improves the speech enhancement performance. Specifically, we propose the VC-S $^{2}$ E method, which incorporates the Conformer and Mamba modules for their complementary strengths. Extensive experiments are conducted on public MUSIC, AVSpeech and AudioSet datasets, where the results demonstrate the superiority of VC-S $^{2}$ E over other competitive methods. We will make the source code publicly available. Project demo page: https://***/ © 2007-2012 IEEE.

关键词： Speech enhancement

来源：评论

学校读者我要写书评

暂无评论

Meta-transfer learning for predicting position-dependent milling stability across varying overhang lengths under limited experimental data

引用

Journal of Manufacturing Processes 2025年 149卷 43-56页

作者： Congying Deng Qiang Wu Jianguo Miao Sheng Lu Luofeng Xie School of Integrated Circuits Chongqing University of Posts and Telecommunications Chongqing 400065 PR China Key Laboratory of Big Data Intelligent Computing Chongqing University of Posts and Telecommunications Chongqing 400065 PR China College of Automation Chongqing University of Posts and Telecommunications Chongqing 400065 PR China School of Mechanical Engineering Sichuan University Chengdu 610065 PR China

Stability lobe diagram (SLD) serves as a pivotal reference in determining appropriate chatter-free milling parameters. The tool tip frequency response functions (FRFs), as crucial prerequisites for deriving theoretical SLDs, vary with the position and overhang length. The acquisition of tool tip FRFs under different machining positions and overhang lengths necessitates repeated time-consuming experiments or simulations. To guarantee the efficiency and accuracy in predicting SLDs, this paper puts forward a meta-transfer learning method utilizing a deep neural network to refine the prediction of stability limits with limited experimental data. Initially, a source overhang length is defined to perform the impact testing under adequate machining positions for acquiring tool tip dynamics and constructing a theoretical stability dataset. Fewer impact tests for a target overhang length are conducted under insufficient machining positions arranged through a proposed similarity-based sampling strategy, and a target stability dataset is constructed using limited tool tip FRFs. Secondly, the source dataset is updated by domain adaptation on the modal parameters with a multi-fidelity modeling. Subsequently, meta-learning is carried out on multiple meta-tasks randomly determined from the updated source dataset. The ultimately trained deep neural network during the meta-learning phase undergoes fine-tuning with the target dataset for predicting the stability limits at various machining positions. Finally, a three-axis vertical machining center is taken to conduct a detailed experimental validation. Comparative analyses indicate that the proposed methodology outperforms others in accurately predicting stability limits, even with a constrained number of labeled samples.

关键词： Milling (machining)

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：