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： Hao, Shugang Duan, Lingjie Pillar of Engineering Systems and Design Singapore University of Technology and Design Singapore

ISBN: (纸本)9798350368741

Continual learning (CL) is a new online learning technique over sequentially generated streaming data from different tasks, aiming to maintain a small forgetting loss on previously-learned tasks. Existing work focuses on reducing the forgetting loss under a given task sequence. However, if similar tasks continuously appear to the end time, the forgetting loss is still huge on prior distinct tasks. In practical IoT networks, an autonomous vehicle to sample data and learn different tasks can route and alter the order of task pattern at increased travelling cost. To our best knowledge, we are the first to study how to opportunistically route the testing object and alter the task sequence in CL. We formulate a new optimization problem and prove it NP-hard. We propose a polynomial-time algorithm to achieve approximation ratios of 3/2 for underparameterized case and 2/3 +r1−T for overparameterized case, respectively. Simulation results verify our algorithm's close-to-optimum performance. © 2025 IEEE.

关键词： approximation algorithm design Continual learning IoT network task ordering

来源：评论

学校读者我要写书评

暂无评论

Online Learning from Strategic Human Feedback in LLM Fine-Tuning

Online Learning from Strategic Human Feedback in LLM Fine-Tu...

引用

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

作者： Hao, Shugang Duan, Lingjie Pillar of Engineering Systems and Design Singapore University of Technology and Design Singapore

ISBN: (纸本)9798350368741

Reinforcement learning from human feedback (RLHF) has become an essential step in fine-tuning large language models (LLMs) to align them with human preferences. However, human labelers are selfish and have diverse preferences. They may strategically misreport their online feedback to influence the system's aggregation towards their own preferences. Current practice simply averages labelers' feedback per time and fails to identify the most accurate human labeler, leading to linear regret O(T) for T time slots. To our best knowledge, we are the first to study online learning mechanisms against strategic human labelers in the LLM fine-tuning process. We formulate a new dynamic Bayesian game and dynamically adjust human labelers' weights in the preference aggregation, ensuring their truthful feedback and sublinear regret O(T1/2). Simulation results demonstrate our mechanism's great advantages over the existing benchmark schemes. © 2025 IEEE.

关键词： LLM fine-tuning online learning regret analysis strategic human feedback truthful mechanism design

来源：评论

学校读者我要写书评

暂无评论

OFDM-structure based waveform designs for integrated sensing and communication

引用

Science China(Information Sciences) 2025年第5期68卷 267-284页

作者： Jiaqi LIU Ping YANG Ke JIANG Jiarong ZHAO Wei XIANG Jianping WEI Ruixiang DUAN Saviour ZAMMIT Tony Q.S.QUEK National Key Laboratory of Wireless Communications University of Electronic Science and Technology of China School of Computing Engineering and Mathematical Sciences La Trobe University Meituan Academy of Robotics Department of Communications and Computer Engineering University of Malta Information Systems Technology and Design Pillar Singapore University of Technology and Design

The future sixth-generation(6G) paradigm aims to seamlessly integrate communication and environmental sensing capabilities into a single radio signal, promising improved efficiency and cost-effectiveness through simultaneous data communications and environmental perception. At the core of this evolution, orthogonal frequency division multiplexing(OFDM) and its advanced waveforms emerge as pivotal for integrated sensing and communications(ISAC). This study introduces a concise and unified ISAC waveform design framework based on orthogonal multicarriers. This framework supports versatile applications of OFDM and its derivative waveforms within a generalized ISAC system, marking a significant leap in integrating communication and sensing capabilities. A distinguishing feature of this framework is its adaptability,allowing users to intelligently select modulation strategies based on their specific environmental needs. This adaptability optimizes performance across diverse scenarios. Central to our innovations is the proposal of discrete Fourier transformspread OFDM with index modulation(DFT-S-OFDM-IM). This framework is paired with newly proposed signal processing methods for single-input single-output and multiple-input multiple-output(MIMO) systems. Extensive evaluations highlight DFT-S-OFDM-IM's superiority, including dramatically reduced peak-to-average power ratios(PAPRs), competitive communication performance, and exceptional sensing capabilities, striking an elegant balance between communication capacity and environmental sensing precision.

关键词： DFT-S-OFDM-IM index modulation integrated sensing and communication OFDM

来源：评论

学校读者我要写书评

暂无评论

Enabling Visual Scene Recovery from Wi-Fi CSI for Occlusion-Free Surveillance

引用

IEEE Internet of Things Journal 2025年第11期12卷 15040-15056页

作者： Chen, Cheng Ohta, Shoki Nishio, Takayuki Bennis, Mehdi Park, Jihong Wahib, Mohamed Tokyo Institute of Technology School of Engineering Meguro-ku Tokyo152-8550 Japan Kobe650-0047 Japan University of Oulu Centre of Wireless Communications Oulu90014 Finland Singapore University of Technology and Design Information Systems Technology and Design Pillar 487372 Singapore

We introduce CSI-Inpainter, a novel approach for obstacle removal using Wi-Fi channel state information (CSI). This method harnesses CSI data to reconstruct obscured visual elements, regardless of lighting conditions. Extensive empirical evaluation in both office and industrial settings demonstrates the effectiveness of CSI-Inpainter's exceptional ability to identify and reconstruct occluded segments, outperforming traditional baselines and our RSSI-based work, RF-Inpainter in terms of visual quality. Our findings emphasize the superiority of CSI data over RSSI for providing richer visual information and underscore the critical role of optimal sensor placement and data fusion from multiple CSI sensors in enhancing the performance. CSI-Inpainter represents a significant advancement in obstacle removal for various applications like surveillance, offering new insights into the integration of wireless sensing and visual scene recovery, expanding the potential applications of Computer Vision in real-world environments. © 2025 IEEE.

关键词： Wi-Fi

来源：评论

学校读者我要写书评

暂无评论

Online Learning from Strategic Human Feedback in LLM Fine-Tuning

Online Learning from Strategic Human Feedback in LLM Fine-Tu...

引用

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

作者： Shugang Hao Lingjie Duan Pillar of Engineering Systems and Design Singapore University of Technology and Design

ISBN: (数字)9798350368741

ISBN: (纸本)9798350368758

Reinforcement learning from human feedback (RLHF) has become an essential step in fine-tuning large language models (LLMs) to align them with human preferences. However, human labelers are selfish and have diverse preferences. They may strategically misreport their online feedback to influence the system’s aggregation towards their own preferences. Current practice simply averages labelers’ feedback per time and fails to identify the most accurate human labeler, leading to linear regret ${\mathcal{O}}(T)$ for T time slots. To our best knowledge, we are the first to study online learning mechanisms against strategic human labelers in the LLM fine-tuning process. We formulate a new dynamic Bayesian game and dynamically adjust human labelers’ weights in the preference aggregation, ensuring their truthful feedback and sublinear regret ${\mathcal{O}}\left({{T^{1/2}}}\right)$. Simulation results demonstrate our mechanism’s great advantages over the existing benchmark schemes.

关键词： Learning systems Simulation Large language models Reinforcement learning Games Benchmark testing Signal processing Acoustics Bayes methods Speech processing

来源：评论

学校读者我要写书评

暂无评论

Algorithm design for Continual Learning in IoT Networks

Algorithm Design for Continual Learning in IoT Networks

引用

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

作者： Shugang Hao Lingjie Duan Pillar of Engineering Systems and Design Singapore University of Technology and Design

ISBN: (数字)9798350368741

ISBN: (纸本)9798350368758

Continual learning (CL) is a new online learning technique over sequentially generated streaming data from different tasks, aiming to maintain a small forgetting loss on previously-learned tasks. Existing work focuses on reducing the forgetting loss under a given task sequence. However, if similar tasks continuously appear to the end time, the forgetting loss is still huge on prior distinct tasks. In practical IoT networks, an autonomous vehicle to sample data and learn different tasks can route and alter the order of task pattern at increased travelling cost. To our best knowledge, we are the first to study how to opportunistically route the testing object and alter the task sequence in CL. We formulate a new optimization problem and prove it NP-hard. We propose a polynomial-time algorithm to achieve approximation ratios of $\frac{3}{2}$ for underparameterized case and $\frac{3}{2} + {r^{1 - T}}$ for overparameterized case, respectively. Simulation results verify our algorithm’s close-to-optimum performance.

关键词： Continuing education Costs Simulation Signal processing algorithms Signal processing Approximation algorithms Polynomials Speech processing Optimization Testing

来源：评论

学校读者我要写书评

暂无评论

To Analyze and Regulate Human-in-the-loop Learning for Congestion Games

arXiv

引用

arXiv 2025年

作者： Li, Hongbo Duan, Lingjie The Pillar of Engineering Systems and Design Singapore University of Technology and Design 487372 Singapore

In congestion games, selfish users behave myopically to crowd to the shortest paths, and the social planner designs mechanisms to regulate such selfish routing through information or payment incentives. However, such mechanism design requires the knowledge of time-varying traffic conditions and it is the users themselves to learn and report past road experiences to the social planner (e.g., Waze or Google Maps). When congestion games meet mobile crowdsourcing, it is critical to incentivize selfish users to explore non-shortest paths in the best exploitation-exploration trade-off. First, we consider a simple but fundamental parallel routing network with one deterministic path and multiple stochastic paths for users with an average arrival probability λ. We prove that the current myopic routing policy (widely used in Waze and Google Maps) misses both exploration (when strong hazard belief) and exploitation (when weak hazard belief) as compared to the social optimum. Due to the myopic policy's under-exploration, we prove that the caused price of anarchy (PoA) is larger than (Equation presented) which can be arbitrarily large as discount factor ρ → 1. To mitigate such huge efficiency loss, we propose a novel selective information disclosure (SID) mechanism: we only reveal the latest traffic information to users when they intend to over-explore stochastic paths upon arrival, while hiding such information when they want to under-explore. We prove that our mechanism successfully reduces PoA to be less than 2. Besides the parallel routing network, we further extend our mechanism and PoA results to any linear path graphs with multiple intermediate nodes. In addition to the worst-case performance evaluation, we conduct extensive simulations with both synthetic and real transportation datasets to demonstrate the close-to-optimal average-case performance of our SID mechanism. © 2025, CC BY.

关键词： Crowdsourcing

来源：评论

学校读者我要写书评

暂无评论

Multi-User Pilot Pattern Optimization for Channel Extrapolation in 5G NR systems

Multi-User Pilot Pattern Optimization for Channel Extrapolat...

引用

IEEE Conference on Wireless Communications and Networking

作者： Yubo Wan An Liu Tony Q. S. Quek College of Information Science and Electronic Engineering Zhejiang University Information Systems Technology and Design Pillar Singapore University of Technology and Design

ISBN: (数字)9798350368369

ISBN: (纸本)9798350368376

Pilot pattern optimization in orthogonal frequency division multiplexing (OFDM) systems has been widely investigated due to its positive impact on channel estimation. In this paper, we consider the problem of multi-user pilot pattern optimization in OFDM systems. In particular, the goal is to enhance channel extrapolation performance for 5G NR systems by optimizing multi-user pilot patterns in frequency-domain. We formulate a novel pilot pattern optimization problem with the objective of minimizing the maximum integrated side-lobe level (ISL) among all users, subject to a statistical resolution limit (SRL) constraint. Unlike existing literature that only utilizes ISL for controlling side-lobe levels of the ambiguity function, we also leverage ISL to mitigate multi-user interference in code-domain. Additionally, the introduced SRL constraint ensures sufficient delay resolution of the system to resolve multipath, thereby improving channel extrapolation performance. Then, we employ the estimation of distribution algorithm (EDA) to solve the formulated problem in an offline manner. Simulation results demonstrate that the optimized pilot pattern yields significant performance gains in channel extrapolation over the conventional pilot patterns.

关键词： Extrapolation 5G mobile communication OFDM Simulation Frequency-domain analysis Estimation Interference Performance gain Delays Optimization

来源：评论

学校读者我要写书评

暂无评论

Competitive Multi-armed Bandit Games for Resource Sharing

引用

IEEE Transactions on Mobile Computing 2025年

作者： Li, Hongbo Duan, Lingjie The Ohio State University Department of Electrical and Computer Engineering ColumbusOH43210 United States Singapore University of Technology and Design Pillar of Engineering Systems and Design 487372 Singapore

In modern resource-sharing systems, multiple agents access limited resources with unknown stochastic conditions to perform tasks. When multiple agents access the same resource (arm) simultaneously, they compete for successful usage, leading to contention and reduced rewards. This motivates our theoretical study of competitive multi-armed bandit (CMAB) games. In this paper, we study a new N-player K-arm competitive MAB game, where non-myopic players (agents) compete with each other to form diverse private estimations of unknown arms over time. Their possible collisions on the same arms and the time-varying nature of arm rewards make the policy analysis here more involved than the existing studies for myopic players. We explicitly analyze the threshold-based structures of the social optimum and the existing selfish policy, showing that the latter causes prolonged convergence times (Formula presented), while the socially optimal policy with coordinated communication reduces it to (Formula presented). Based on the policy comparison, we prove that the competition among selfish players for the best arm can result in an infinite price of anarchy (PoA), indicating an arbitrarily large efficiency loss compared to the social optimum. We further prove that no informational (non-monetary) mechanism (including Bayesian persuasion) can reduce the infinite PoA, as strategic misreporting by non-myopic players undermines such approaches. To address this, we propose a Combined Informational and Side-Payment (CISP) mechanism, which provides socially optimal arm recommendations with proper informational and monetary incentives to players according to their diverse and time-varying private beliefs. Our CISP mechanism keeps ex-post budget balanced for the social planner and ensures truthful reporting from players, thereby achieving the minimum PoA = 1 and the same convergence time as the social optimum. © 2002-2012 IEEE.

关键词： Budget control

来源：评论

学校读者我要写书评

暂无评论

EraW-Net: Enhance-Refine-Align W-Net for Scene-Associated Driver Attention Estimation

引用

IEEE Transactions on Multimedia 2025年

作者： Zhou, Jun Liu, Chunsheng Chang, Faliang Wang, Wenqian Hao, Penghui Huang, Yiming Yang, Zhiqiang Shandong University School of Control Science and Engineering Jinan250061 China Singapore University of Technology and Design Pillar of Information Systems Technology and Design 487372 Singapore

Associating driver attention with driving scene across two fields of view is a challenging cross-domain perception problem, which requires comprehensive consideration of cross-view mapping, dynamic driving scene analysis and driver status tracking. Previous methods typically analyze a single view or map attention to the scene through a two-step projection, failing to exploit their implicit connections and establish accurate associations. Moreover, simple fusion modules are inadequate for modeling the complex relationships between the two views, making information integration complicated. To address these issues, we propose EraW-Net, a novel end-to-end framework for scene-associated driver attention estimation by aggregating information from dual views. This method enhances the most discriminative dynamic cues, refines feature representations, and facilitates semantically aligned cross-domain integration through a W-shaped architecture, termed W-Net. Specifically, a Dynamic Adaptive Filter Module (DAF-Module) is proposed to address the challenges of frequently changing driving environments by extracting vital regions. It suppresses the indiscriminately recorded dynamics and highlights crucial ones by innovative joint frequency-spatial analysis, enhancing the model's ability to parse complex dynamics. Additionally, to track driver states during non-fixed facial poses, we propose a Global Context Sharing Module (GCS-Module) to construct refined feature representations by capturing hierarchical features that adapt to various scales of head and eye movements. Finally, W-Net achieves systematic cross-view information integration through its unique two-stage decoding strategy, addressing semantic misalignment in heterogeneous data integration. Experiments demonstrate that the proposed method robustly and accurately estimates scene-associated driver attention on large public datasets. © 1999-2012 IEEE.

关键词： Mapping

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：