检索结果-内蒙古大学图书馆

2024 Findings of the Association for Computational Linguistics: NAACL 2024

作者： Li, Wendi Wei, Wei Xu, Kaihe Xie, Wenfeng Chen, Dangyang Cheng, Yu Cognitive Computing and Intelligent Information Processing Laboratory School of Computer Science and Technology Huazhong University of Science and Technology China China Ping An Property & Casualty Insurance Company of China China The Chinese University of Hong Kong Hong Kong

ISBN: (纸本)9798891761193

To meet the requirements of real-world applications, it is essential to control generations of large language models (LLMs). Prior research has tried to introduce reinforcement learning (RL) into controllable text generation while most existing methods suffer from overfitting issues (finetuning-based methods) or semantic collapse (post-processing methods). However, current RL methods are generally guided by coarse-grained (sentence/paragraph-level) feedback, which may lead to suboptimal performance owing to semantic twists or progressions within sentences. To tackle that, we propose a novel reinforcement learning algorithm named TOLE which formulates TOken-LEvel rewards for controllable text generation, and employs a "first-quantize-then-noise" paradigm to enhance the robustness of the RL algorithm. Furthermore, TOLE can be flexibly extended to multiple constraints with little computational expense. Experimental results show that our algorithm can achieve superior performance on both single-attribute and multi-attribute control tasks. We have released our codes at https://***/WindyLee0822/CTG. © 2024 Association for Computational Linguistics.

关键词： Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Policy iteration-based adaptive optimal control for Markov jump systems: a transition-probability-free asynchronous approach

引用

Science China(information Sciences) 2024年第11期67卷 363-364页

作者： Weidi CHENG Chengcheng REN Shuping HE Changyin SUN Key Laboratory of Intelligent Computing and Signal Processing (Ministry of Education) School of Electrical Engineering and Automation Anhui University Anhui Engineering Laboratory of Human-Robot Integration System and Intelligent Equipment School of Electrical Engineering and Automation Anhui University Information Materials and Intelligent Sensing Laboratory of Anhui Province Engineering Research Center of Autonomous Unmanned System Technology (Ministry of Education) School of Artificial Intelligence Anhui University

The controller and filter design problems of Markov jump systems(MJSs) have gained significant attention over the past few decades. These studies include various aspects,including stochastic stabilization [1], optimal tracking control [2], and dissipative filter design [3]. Although numerous publications address the optimal controller design for MJSs, the issue of hidden MJSs, particularly those with mismatched jumping modes between the system and the controller, has rarely been explored.

关键词： Markov processes

来源：评论

学校读者我要写书评

暂无评论

Optical and Electrical Analysis of an Advanced 4T All-perovskite Tandem Solar Cell

Optical and Electrical Analysis of an Advanced 4T All-perovs...

引用

2024 Photonics and Electromagnetics Research Symposium, PIERS 2024

作者： Xia, Zhaosheng Ren, Xingang Wei, Shengyang Huang, Zhixiang Anhui University Information Materials and Intelligent Sensing Laboratory of Anhui Province Institute of Physical Science and Information Technology Hefei230039 China Anhui University Key Laboratory of Intelligent Computing and Signal Processing Ministry of Education Hefei230601 China

ISBN: (纸本)9798350375909

The investigation of the optoelectronic characteristics of all-perovskite tandem solar cells holds pivotal significance in surpassing the Shockley-Queisser limit of single-junction perovskite solar cells. Initially, we established two distinct four-terminal tandem solar cells (4T TSCs) composed of different perovskite materials: FA0.8Cs0.2Pb(I0.7Br0.3)3 (Eg = 1.75 eV) and MASnI3 (Eg = 1.3 eV). Through numerical simulations of the optoelectronic characteristics of 4T TSCs, we evaluated the performance of the TSCs at varying perovskite layer thicknesses. Following optimization, the power conversion efficiency (PCE) of the 4T TSCs increased from the initial 30.34% to 33.54%. Through an in-depth analysis of the cell and energy distribution in TSCs, it was observed that the light-coupling layer significantly influenced the PCE of the 4T TSCs. Upon optimizing the light-coupling layer, the PCE of the TSC increased from the initial 30.34% to 34.7%. This study provides crucial theoretical and experimental foundations for optimizing the performance of all-perovskite TSCs. © 2024 IEEE.

关键词： Perovskite solar cells

来源：评论

学校读者我要写书评

暂无评论

Network link weight optimization based on antisymmetric deep graph networks and reinforcement learning 6

Network link weight optimization based on antisymmetric deep...

引用

6th International Conference on Next Generation Data-Driven Networks, NGDN 2024

作者： Chen, Yuhan Guo, Yingya Department of Computer and Data Science Fuzhou University Fuzhou China Key Laboratory of Network Computing and Intelligent Information Processing Fuzhou University Fuzhou China

ISBN: (纸本)9798350388374

Route optimization is a key core technology to optimize network traffic distribution, achieve network load balancing, and improve network performance. Traditional distributed networks widely run shortest-path based routing protocols, and the path of traffic is determined by the link weights of the distributed network, so routing optimization methods are usually optimized around the link weights of the network. Heuristics-based link weight optimization methods are widely used. However, heuristic methods rely on manually set rules, which are poorly generalized and cannot adapt well to dynamically changing traffic demands. Compared to heuristics, deep reinforcement learning (DRL) has the advantage of extracting more accurate feature representations in addition to its ability to handle high-dimensional state and action spaces. We propose a network link weight optimization method based on anti-symmetric deep graph networks (A-DGN) and reinforcement learning using a novel GNN framework anti-symmetric deep graph networks, where link weights are adjusted to reduce the network link utilization with the optimization objective of minimizing the maximum link utilization in the network. Experimental results show that the proposed method achieves significant performance improvements in the link weight optimization problem in four real-world network topology scenarios. © 2024 IEEE.

关键词： Deep reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Lightweight Food Image Recognition With Global Shuffle Convolution

IEEE Transactions on AgriFood Electronics

引用

IEEE Transactions on AgriFood Electronics 2024年第2期2卷 392-402页

作者： Sheng, Guorui Min, Weiqing Yao, Tao Song, Jingru Yang, Yancun Wang, Lili Jiang, Shuqiang Ludong University Department of Information and Electrical Engineering Yantai264025 China Institute of Computing Technology Chinese Academy of Sciences Key Laboratory of Intelligent Information Processing Beijing100190 China University of Chinese Academy of Sciences Beijing100049 China

Consumer behaviors and habits in food choices impact their physical health and have implications for climate change and global warming. Efficient food image recognition can assist individuals in making more environmentally friendly and healthier dietary choices using end devices, such as smartphones. Simultaneously, it can enhance the efficiency of server-side training, thereby reducing carbon emissions. We propose a lightweight deep neural network named Global Shuffle Net (GSNet) that can efficiently recognize food images. In GSNet, we develop a novel convolution method called global shuffle convolution, which captures the dependence between long-range pixels. Merging global shuffle convolution with classic local convolution yields a framework that works as the backbone of GSNet. Through GSNet's ability to capture the dependence between long-range pixels at the start of the network, by restricting the number of layers in the middle and rear, the parameters and floating operation operations (FLOPs) can be minimized without compromising the performance, thus permitting a lightweight goal to be achieved. Experimental results on four popular food recognition datasets demonstrate that our approach achieves state-of-the-art performance with higher accuracy and fewer FLOPs and parameters. For example, in comparison to the current state-of-the-art model of MobileViTv2, GSNet achieved 87.9% accuracy of the top-1 level on the Eidgenössische Technische Hochschule Zürich (ETHZ) Food-101 dataset with 28% reduction in the parameters, 37% reduction in the FLOPs, but a 0.7% more accuracy. © 2024 IEEE.

关键词： Carbon dioxide

来源：评论

学校读者我要写书评

暂无评论

Pseudo Multi-Source Domain Extension and Selective Pseudo-Labeling for Unsupervised Domain Adaptive Medical Image Segmentation 48

Pseudo Multi-Source Domain Extension and Selective Pseudo-La...

引用

48th IEEE International Conference on Acoustics, Speech and Signal processing, ICASSP 2023

作者： Liu, Xiaokang Wang, Zhiqiang Hu, Kai Gao, Xieping Xiangtan University Key Laboratory of Intelligent Computing and Information Processing of Ministry of Education Xiangtan411105 China Xiangnan University College of Medical Imaging Laboratory and Rehabilitation Chenzhou423000 China Hunan Normal University Hunan Provincial Key Laboratory of Intelligent Computing and Language Information Processing Changsha410081 China

ISBN: (纸本)9781728163277

Unsupervised domain adaptation (UDA) attracts extra attention in medical image processing because no additional labels are required when adapting to different distributions. In this work, we propose a novel unsupervised domain adaptation framework named as Domain Expansion and PseudoLabeling (DEPL). We extend the domain of a labeled source domain data to four different distributed domains and use adversarial learning to align the image appearance level and feature level from the four different domains to the unlabeled target domain. In addition, we propose a selective pseudolabeling mechanism, namely using strong confidence pseudolabeling to boost model performance. We evaluate our model for the MR to CT adaptation segmentation task on the public dataset MMWHS. Compared to seven other state-of-the-art segmentation methods, our DEPL achieves the best Dice similarity coefficient by 82.4%, which is at least 3.9% higher than the other UDA segmentation methods. © 2023 IEEE.

关键词： Adversarial learning domain extension Image segmentation Selective pseudo-labeling Unsupervised domain adaptive

来源：评论

学校读者我要写书评

暂无评论

Scheduling optimization for upstream dataflows in edge computing

引用

Digital Communications and Networks 2023年第6期9卷 1448-1457页

作者： Haohao Wang Mengmeng Sun Lianming Zhang Pingping Dong Yehua Wei Jing Mei College of Information Science and Engineering Hunan Normal UniversityChangsha410081China Hunan Provincial Key Laboratory of Intelligent Computing and Language Information Processing Hunan Normal UniversityChangsha410081China Hunan Xiangjiang Artificial Intelligence Academy Hunan Normal UniversityChangsha410081China

Edge computing can alleviate the problem of insufficient computational resources for the user equipment,improve the network processing environment,and promote the user *** computing is well known as a prospective method for the development of the Internet of Things(IoT).However,with the development of smart terminals,much more time is required for scheduling the terminal high-intensity upstream dataflow in the edge server than for scheduling that in the downstream *** this paper,we study the scheduling strategy for upstream dataflows in edge computing networks and introduce a three-tier edge computing network *** propose a Time-Slicing Self-Adaptive Scheduling(TSAS)algorithm based on the hierarchical queue,which can reduce the queuing delay of the dataflow,improve the timeliness of dataflow processing and achieve an efficient and reasonable performance of dataflow *** experimental results show that the TSAS algorithm can reduce latency,minimize energy consumption,and increase system throughput.

关键词： Edge computing Time-slicing Dataflow scheduling Dynamic analysis

来源：评论

学校读者我要写书评

暂无评论

Few-Shot Class-Incremental Semantic Segmentation via Pseudo-Labeling and Knowledge Distillation 4

Few-Shot Class-Incremental Semantic Segmentation via Pseudo-...

引用

4th International Conference on information Science, Parallel and Distributed Systems, ISPDS 2023

作者： Jiang, Chengjia Wang, Tao Li, Sien Wang, Jinyang Wang, Shirui Antoniou, Antonios Minjiang University Fujian Provincial Key Laboratory of Information Processing and Intelligent Control Fuzhou China Fujian Education Institutions Wuyi University The Key Laboratory of Cognitive Computing and Intelligent Information Processing Wuyishan China College of Computer and Data Science Fuzhou University Fuzhou China European University Cyprus Department of Computer Science and Engineering Nicosia Cyprus

ISBN: (纸本)9798350337181

We address the problem of learning new classes for semantic segmentation models from few examples, which is challenging because of the following two reasons. Firstly, it is difficult to learn from limited novel data to capture the underlying class distribution. Secondly, it is challenging to retain knowledge for existing classes and to avoid catastrophic forgetting. For learning from limited data, we propose a pseudo-labeling strategy to augment the few-shot training annotations in order to learn novel classes more effectively. Given only one or a few images labeled with the novel classes and a much larger set of unlabeled images, we transfer the knowledge from labeled images to unlabeled images with a coarse-to-fine pseudo-labeling approach in two steps. Specifically, we first match each labeled image to its nearest neighbors in the unlabeled image set at the scene level, in order to obtain images with a similar scene layout. This is followed by obtaining pseudo-labels within this neighborhood by applying classifiers learned on the few-shot annotations. In addition, we use knowledge distillation on both labeled and unlabeled data to retain knowledge on existing classes. We integrate the above steps into a single convolutional neural network with a unified learning objective. Extensive experiments on the Cityscapes and KITTI datasets validate the efficacy of the proposed approach in the self-driving domain. Code is available from https://***/ChasonJiang/FSCILSS. © 2023 IEEE.

关键词： Distillation

来源：评论

学校读者我要写书评

暂无评论

Open-vocabulary 3D Semantic Understanding via Affinity Neural Radiance Fields 6

Open-vocabulary 3D Semantic Understanding via Affinity Neura...

引用

6th International Conference on Next Generation Data-Driven Networks, NGDN 2024

作者： Fan, Yujie Luo, Huan Computer and Data Science Fuzhou University Fuzhou China Fujian Provincial Key Laboratory of Network Computing and Intelligent Information Processing Fuzhou University Fuzhou China

ISBN: (纸本)9798350388374

Three-dimensional semantic understanding using only several multi-view images can largely reduce the communication burden on the network. In addition, while point clouds are extensively studied for 3D scene understanding, utilization of multi-view image data offers rich visual details and texture information. However, challenges persist in lifting 2D semantic features to 3D space and leveraging language for segmentation. Inspired by recent advancements, this paper proposes a method that combines CLIP features and SAM masks to create a feature field capable of segmenting objects via natural language text across 2D multi-view and 3D Gaussian splatting. It offers a promising function for extracting 3D assets for game engines and the metaverse. Our method involves mask generation from video frames, extracting physical scales via RGB Nerf with masks, and organizing hierarchical information for semantic comprehension. In the training process, affinity features maintain scale properties and guide CLIP feature generation with auto weights blending for semantic robustness. A straightforward 3D splatting CLIP feature approach and canonical text methodology enhance query robustness across 2D multi-view and 3D splatting through relevance score calculation based on text CLIP features for inference. Experimental results demonstrate promising improvements in semantic understanding of 3D scenes. © 2024 IEEE.

关键词： Semantic Segmentation

来源：评论

学校读者我要写书评

暂无评论

Linear Precoding for Faster-Than-Nyquist Signaling with Imperfect Channel State information 16

Linear Precoding for Faster-Than-Nyquist Signaling with Impe...

引用

16th International Conference on Wireless Communications and Signal processing, WCSP 2024

作者： Wang, Yan Li, Qiang Li, Yingsong Li, Liping Li, Xingwang Anhui University Key Laboratory of Intelligent Computing and Signal Processing Ministry of Education China School of Physics and Electronic Information Engineering Henan Polytechnic University China

ISBN: (纸本)9798350390643

In faster-than-Nyquist (FTN) transmission over frequency-selective fading channels, it is essential to eliminate both FTN-induced intersymbol-interference (ISI) (FTN-ISI) and channel-induced ISI (channel-ISI). Although frequency domain equalization (FDE) techniques can mitigate channel-ISI at the receiver, their performance is often inadequate. In contrast, cyclic prefix/suffix (CPS)-G-to-minus-half (GTMH) precoding shows superior effectiveness in mitigating channel-ISI. However, existing CPS-GTMH precoding techniques assume the transmitter obtains perfect channel state information (CSI) estimated at the receiver. This is impractical due to the high latency in satellite communication links. Therefore, this paper explores precoding for FTN signaling under imperfect CSI (ICSI) and proposes CPS-minimum mean squared error (MMSE) and linear pre-equalization (LPE)-MMSE precoding schemes based on the MMSE criterion. Both schemes employ a cyclic prefix to transform the channel-ISI matrix into the circular matrix, thereby avoiding inter-block interference. The primary distinction is that CPS-MMSE precoding handles FTN-ISI and channel-ISI together using MMSE precoding. In contrast, LPE-MMSE precoding addresses FTN-ISI with LPE and channel-ISI with MMSE precoding separately. Simulation results indicate that CPS-MMSE and LPE-MMSE precodings outperform CPS-FDE and LPE-FDE in mitigating channel-ISI when the estimation error is 0.005 and 0.01. For instance, with a packing ratio of 0.9, a roll-off factor of 0.25, and an estimation error of 0.01, CPS-MMSE and LPE-MMSE precodings achieve bit error rate (BER) gains of approximately 0.8 dB and 1 dB at a BER of 1e4, respectively. However, when the estimation error increases to 0.1, the BER performance of CPS-FDE and LPE-FDE surpasses that of the CPS-MMSE and LPE-MMSE precodings, indicating the proposed precoding schemes' unsuitability for larger channel estimation errors. © 2024 IEEE.

关键词： Bit error rate

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：