检索结果-内蒙古大学图书馆

Centroid-based clustering validity: method and application to quantification of optimal cluster-data space

Soft Computing 2024年第19期28卷 10853-10872页

作者： Nguyen, Sy Dzung Laboratory for Computational Mechatronics Institute for Computational Science and Artificial Intelligence Van Lang University Ho Chi Minh Viet Nam Faculty of Mechanical - Electrical and Computer Engineering School of Technology Van Lang University Ho Chi Minh Viet Nam

Evaluation of clustering validity to set up an optimal cluster-data space (CDS) is a vital task in many fields related to data mining. Almost existing clustering validity indexes (CVIs) lack stability due to being too sensitive to noise, especially impulse noise. Here, we (1) propose a new CVI named DzI (Dzung Index) or fRisk2 using analysis of fuzzy-set-based accumulated risk degree (FARD), and (2) present a new algorithm named fRisk2-bA for determining the optimal number of data clusters. It is a method of evaluation of the centroid-based fuzzy clustering validity. In essence, the fRisk2 still focuses on enhancing the data compression in each cluster and expanding the separation between cluster centroids. However, these features are exploited indirectly through FARD. As a result, the proposed method not only can avoid the difficulties of the traditional ones relying on the compression and separation properties directly but also can distill better local and global attributes in the data distribution to estimate the CDS more fully. Along with the proved theory basis, surveys, including the ones based on noisy datasets from measurements, showed the compared advantages of fRisk2 as follows. (1) The accuracy, stability, and convergence of the fRisk2 are outstanding. (2) Its total calculating cost is lower than the other surveyed CVIs. © The Author(s), under exclusive licence to Springer-Verlag GmbH Germany, part of Springer Nature 2024.

关键词： Fuzzy clustering

来源：评论

学校读者我要写书评

暂无评论

Adapted Ant Colony Optimization for Team Orienteering Problem 11

Adapted Ant Colony Optimization for Team Orienteering Proble...

引用

11th International Conference on Machine intelligence Theory and Applications, MiTA 2024

作者： Wu, De-Ming Duan, Dan-Ting Yang, Qiang Liu, Xiao-Fang Zhou, Chang-Jun Zhao, Jian-Min Zhejiang Normal University School of Computer Science and Technology Jinhua China Communication University of China Key Laboratory of Media Audio & Video Beijing China Nanjing University of Information Science and Technology School of Artificial Intelligence Nanjing China Nankai University College Of Artificial Intelligence Tianjin China

ISBN: (纸本)9798350385670

The Team Orienteering Problem (TOP) is to find multiple optimal subsets of nodes and the associated routes to visit these nodes with the objective to maximize the total profits and under the constraint that the traveling cost of each route does not exceed the given cost limit. Such a problem is very challenging and difficult to solve because it involves node selection optimization, route optimization, and cost constraints for multiple routes. Thanks to the significantly good performance of Ant Colony Optimization (ACO) algorithms in solving path planning problems, they are considered promising for addressing TOP. To this end, this paper takes the first attempt to adapt five classical ACO algorithms, namely the Ant System (AS), Elitist Ant System (EAS), Rank-based Ant System (AS rank), Min-Max Ant System (MMAS), and Ant Colony System (ACS), to solve TOP. To achieve this, we maintain an ant team consisting of multiple ants to construct a feasible solution containing multiple routes. Further, we design an ant selection scheme for the ant team to select the next node to visit in the route construction by considering the given cost constraints. To assess the optimization performance of the five adapted ACO algorithms in solving TOP, experiments are conducted on the classical TOP instances. The results reveal that all five ACO algorithms are promising for addressing TOP. Particularly, ACS excels in maximizing the overall team profit, while EAS demonstrates good stability in minimizing the profit differences between routes. © 2024 IEEE.

关键词： Ant colony optimization

来源：评论

学校读者我要写书评

暂无评论

ADAPTIVE STOCHASTIC GRADIENT ALGORITHM FOR BLACK-BOX MULTI-OBJECTIVE LEARNING 12

ADAPTIVE STOCHASTIC GRADIENT ALGORITHM FOR BLACK-BOX MULTI-O...

引用

12th International Conference on Learning Representations, ICLR 2024

作者： Ye, Feiyang Lyu, Yueming Wang, Xuehao Zhang, Yu Tsang, Ivor W. Department of Computer Science and Engineering Southern University of Science and Technology China Australian Artificial Intelligence Institute University of Technology Sydney Australia Centre for Frontier AI Research Agency for Science Technology and Research Singapore Institute of High Performance Computing Agency for Science Technology and Research Singapore Shanghai Artificial Intelligence Laboratory China School of Computer Science and Engineering Nanyang Technological University Singapore

Multi-objective optimization (MOO) has become an influential framework for various machine learning problems, including reinforcement learning and multitask learning. In this paper, we study the black-box multi-objective optimization problem, where we aim to optimize multiple potentially conflicting objectives with function queries only. To address this challenging problem and find a Pareto optimal solution or the Pareto stationary solution, we propose a novel adaptive stochastic gradient algorithm for black-box MOO, called ASMG. Specifically, we use the stochastic gradient approximation method to obtain the gradient for the distribution parameters of the Gaussian smoothed MOO with function queries only. Subsequently, an adaptive weight is employed to aggregate all stochastic gradients to optimize all objective functions effectively. Theoretically, we explicitly provide the connection between the original MOO problem and the corresponding Gaussian smoothed MOO problem and prove the convergence rate for the proposed ASMG algorithm in both convex and non-convex scenarios. Empirically, the proposed ASMG method achieves competitive performance on multiple numerical benchmark problems. Additionally, the state-of-the-art performance on the black-box multi-task learning problem demonstrates the effectiveness of the proposed ASMG method. © 2024 12th International Conference on Learning Representations, ICLR 2024. All rights reserved.

关键词： Multiobjective optimization

来源：评论

学校读者我要写书评

暂无评论

GRU-Based Winner Subcarrier Detection in Frequency Domain Contention 24

GRU-Based Winner Subcarrier Detection in Frequency Domain Co...

引用

7th International Conference on Information science and Systems, ICISS 2024

作者： Yao, Shumin Zhao, Qinglin Xu, Zezhou Feng, Li Li, Guanghui Tian, Liwei Sun, Yi School of Computer Science and Engineering Macau University of Science and Technology China Department of Broadband Communication Pengcheng Laboratory China School of Artificial Intelligence and Computer Science Jiangnan University China Department of Computer Guangdong University of Science and Technology China Institute of Computing Technology Chinese Academy of Sciences China

ISBN: (纸本)9798400717567

Single-carrier frequency domain contention (S-FDC) is an efficient wireless contention mechanism based on orthogonal frequency-division multiplexing (OFDM). In each round of S-FDC, each node randomly selects and signals on one of the OFDM subcarriers while listening to all subcarriers simultaneously. Each node independently detects the activated subcarrier with the smallest index and regards it as the winner subcarrier. Accurately detecting the winner subcarrier is crucial to ensuring the functionality of the network. However, it is extremely difficult in practice due to power leakage problems that arise from frequency asynchronization among nodes. In this paper, we propose a gated-recurrent-unit (GRU)-based model that incorporates S-FDC’s unique features as domain knowledge to enable accurate winner subcarrier detection. Extensive simulation experiments verify that our DL model is effective and outperforms state of the art. This research highlights the potential of DL in improving wireless communication systems. © 2024 Copyright held by the owner/author(s).

关键词： Frequency domain analysis

来源：评论

学校读者我要写书评

暂无评论

SpikeReveal: Unlocking Temporal Sequences from Real Blurry Inputs with Spike Streams 38

SpikeReveal: Unlocking Temporal Sequences from Real Blurry I...

引用

38th Conference on Neural Information Processing Systems, NeurIPS 2024

作者： Chen, Kang Chen, Shiyan Zhang, Jiyuan Zhang, Baoyue Zheng, Yajing Huang, Tiejun Yu, Zhaofei School of Computer Science Peking University China State Key Laboratory for Multimedia Information Processing Peking University China Institute for Artificial Intelligence Peking University China

Reconstructing a sequence of sharp images from the blurry input is crucial for enhancing our insights into the captured scene and poses a significant challenge due to the limited temporal features embedded in the blurry image. Spike cameras, sampling at rates up to 40,000 Hz, have proven effective in capturing motion features and beneficial for solving this ill-posed problem. Nonetheless, existing methods fall into the supervised learning paradigm, which suffers from notable performance degradation when applied to real-world scenarios that diverge from the synthetic training data domain. To address this challenge, we propose the first self-supervised framework for the task of spike-guided motion deblurring. Our approach begins with the formulation of a spike-guided deblurring model that explores the theoretical relationships among spike streams, blurry images, and their corresponding sharp sequences. We subsequently develop a self-supervised cascaded framework to alleviate the issues of spike noise and spatial-resolution mismatching encountered in the deblurring model. With knowledge distillation and reblur loss, we further design a lightweight deblur network to restore high-quality sequences with brightness and texture consistency with the original input. Quantitative and qualitative experiments conducted on our real-world and synthetic datasets with spikes validate the superior generalization of the proposed framework. Our code, data and trained models are available at https://***/chenkang455/S-SDM. © 2024 Neural information processing systems foundation. All rights reserved.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Hypernetwork-Assisted Parameter-Efficient Fine-Tuning with Meta-Knowledge Distillation for Domain Knowledge Disentanglement

Hypernetwork-Assisted Parameter-Efficient Fine-Tuning with M...

引用

2024 Findings of the Association for Computational Linguistics: NAACL 2024

作者： Li, Changqun Wang, Linlin Lin, Xin Huang, Shizhou He, Liang School of Computer Science and Technology East China Normal University China Shanghai Key Laboraiory of Multdimensional Infomation Procesing China Shanghai Artificial Intelligence Laboratory China

ISBN: (纸本)9798891761193

Domain adaptation from labeled source domains to the target domain is important in practical summarization scenarios. However, the key challenge is domain knowledge disentanglement. In this work, we explore how to disentangle domain-invariant knowledge from source domains while learning specific knowledge of the target domain. Specifically, we propose a hypernetwork-assisted encoder-decoder architecture with parameter-efficient fine-tuning. It leverages a hypernetwork instruction learning module to generate domain-specific parameters from the encoded inputs accompanied by task-related instruction. Further, to better disentangle and transfer knowledge from source domains to the target domain, we introduce a meta-knowledge distillation strategy to build a meta-teacher model that captures domain-invariant knowledge across multiple domains and use it to transfer knowledge to students. Experiments on three dialogue summarization datasets show the effectiveness of the proposed model. Human evaluations also show the superiority of our model with regard to the summary generation quality. © 2024 Association for Computational Linguistics.

关键词： Distillation

来源：评论

学校读者我要写书评

暂无评论

VulD-CodeBERT: CodeBERT-Based Vulnerability Detection Model for C/C++ Code 6

VulD-CodeBERT: CodeBERT-Based Vulnerability Detection Model ...

引用

6th International Conference on Communications, Information System and computer Engineering, CISCE 2024

作者： Xiong, Zhe Dong, Weiyu School of Computer Science and Artificial Intelligence Zhengzhou University Zhengzhou China Information Engineering University State Key Laboratory of Mathematical Engineering and Advanced Computing Zhengzhou China

ISBN: (纸本)9798350353174

Recently, deep learning techniques have received widespread attention in code vulnerability and vulnerability detection. However, there are some problems with current vulnerability detection techniques, firstly, because of the non-homogeneity of vulnerability categories and quantities, there are omission false positives and overfitting problems for vulnerabilities with a small amount of data. Second, the existing detection methods have limited ability to characterize the code, there is insufficient depth of characterization, does not contain all of its vulnerability characteristics or is interspersed with useless information. Finally, the current research does not use the powerful feature mining capability of pre-trained models and does not employ model fusion to increase the effectiveness of model vulnerability detection. To solve the above problems, this paper proposes a new deep learning based pre-trained model for code vulnerability detection tool VulD-CodeBERT. By fine-tuning CodeBERT, a code adaptation extension for pre-trained models developed by Microsoft, the CodeBERT model can solve the problem of not being able to learn the vulnerability patterns due to insufficient data, thus resulting in missed false alarms. The sequence output from the CodeBERT model are added to a BiLSTM network with an attention mechanism to mine the local key features of the vulnerabilities combined with the global features of the pooler output from the CodeBERT model to deeply mine the dependencies between the vulnerability statements and derive the probability of the existence of the vulnerability. For vulnerability types that are difficult to uncover captured features or the number of samples is not enough, a polynomial plain Bayesian classifier is used for detection, and finally the two models are combined to deeply mine the vulnerability features, thus solving the problem of difficult and small samples, and reducing the leakage rate and false alarm rate. Experimental results sh

关键词： C++ (programming language)

来源：评论

学校读者我要写书评

暂无评论

ToolRerank: Adaptive and Hierarchy-Aware Reranking for Tool Retrieval 30

ToolRerank: Adaptive and Hierarchy-Aware Reranking for Tool ...

引用

Joint 30th International Conference on Computational Linguistics and 14th International Conference on Language Resources and Evaluation, LREC-COLING 2024

作者： Zheng, Yuanhang Li, Peng Liu, Wei Liu, Yang Luan, Jian Wang, Bin Department of Computer Science and Technology Tsinghua University Beijing China Tsinghua University Beijing China Shanghai Artificial Intelligence Laboratory Shanghai China Xiaomi AI Lab

ISBN: (纸本)9782493814104

Tool learning aims to extend the capabilities of large language models (LLMs) with external tools. A major challenge in tool learning is how to support a large number of tools, including unseen tools. To address this challenge, previous studies have proposed retrieving suitable tools for the LLM based on the user query. However, previously proposed methods do not consider the differences between seen and unseen tools, nor do they take the hierarchy of the tool library into account, which may lead to suboptimal performance for tool retrieval. Therefore, to address the aforementioned issues, we propose ToolRerank, an adaptive and hierarchy-aware reranking method for tool retrieval to further refine the retrieval results. Specifically, our proposed ToolRerank includes Adaptive Truncation, which truncates the retrieval results related to seen and unseen tools at different positions, and Hierarchy-Aware Reranking, which makes retrieval results more concentrated for single-tool queries and more diverse for multi-tool queries. Experimental results show that ToolRerank can improve the quality of the retrieval results, leading to better execution results generated by the LLM. © 2024 ELRA Language Resource Association: CC BY-NC 4.0.

关键词： Computational linguistics

来源：评论

学校读者我要写书评

暂无评论

Giant saturation absorption of tungsten trioxide film prepared based on the seedless layer hydrothermal method

引用

Chinese Physics B 2023年第3期32卷 286-290页

作者：马晓光胡芳珍陈希王艺盟郝晓剑顾敏张启明 Science and Technology on Electronic Test and Measurement Laboratory North University of ChinaTaiyuan 030051China Institute of Photonic Chips University of Shanghai for Science and TechnologyShanghai 200093China Centre for Artificial-Intelligence Nanophotonics School of Optical-Electrical and Computer EngineeringUniversity of Shanghai for Science and TechnologyShanghai 200093China

Nonlinear materials have gained wide interest as saturable absorbers and pulse compression for pulsed laser applications due to their unique optical *** work investigates the third-order nonlinear phenomenon of tungsten trioxide(WO_(3))thin *** giant nonlinear absorption and nonlinear refractive index of WO_(3)thin films were characterized by Z-scan method at 800 *** experimentally observed the giant saturable absorption(SA)and nonlinear refractive index of WO_(3)thin films prepared by the seedless layer hydrothermal method,with SA coefficient being as high as-2.59×105cm·GW^(-1).The SA coefficient is at least one order of magnitude larger than those of the conventional *** nonlinear refractive index n_(2)of WO_(3)film has been observed for the first time in recent studies and the corresponding coefficient can be up to 1.793 cm^(2)·GW^(-1).The large third-order nonlinear optical(NLO)response enables WO_(3)thin films to be promising candidates for optoelectronic and photonic applications in the near-infrared domain.

关键词： tungsten trioxide Z-scan saturable absorption

来源：评论

学校读者我要写书评

暂无评论

TRAP: Tailored Robustness Assessment for Index Advisors via Adversarial Perturbation 40

TRAP: Tailored Robustness Assessment for Index Advisors via ...

引用

40th IEEE International Conference on Data Engineering, ICDE 2024

作者： Zhou, Wei Lin, Chen Zhou, Xuanhe Li, Guoliang Wang, Tianqing School of Informatics Shanghai Artificial Intelligence Laboratory Xiamen University China Tsinghua University Dept. of Computer Science Beijing China Huawei Company Beijing China

ISBN: (纸本)9798350317152

Many index advisors have recently been proposed to build indexes automatically to improve query performance. However, they mainly consider performance improvement in static scenarios. Their robustness, i.e., stable performance in dynamic scenarios (e.g., with minor workload changes), has not been well investigated. This paper addresses the challenges of assessing the index advisor's robustness from the following aspects. First, we introduce perturbation-based workloads for robustness assessment and identify three typical perturbation constraints that occur in real scenarios. Second, with the perturbation constraints, we formulate the generation of perturbed queries as a sequence-to-sequence problem and propose TRAP (Tailored Robustness assessment via Adversarial Perturbation) to pinpoint the performance loopholes of index advisors. Third, to generalize to various index advisors, we place TRAP in an opaque-box setting (i.e., with little knowledge of the index advisors' internal design), and we propose a two-phase training paradigm to efficiently train TRAP without elaborately annotated data. Fourth, we conduct comprehensive robustness assessments on standard benchmarks and real workloads for ten existing index advisors. Our findings reveal that these index advisors are vulnerable to the workloads generated by TRAP. Finally, based on the assessment results, we shed light on insights to enhance the robustness of different index advisors. For example, learning-based index advisors can benefit from adopting a fine-grained state representation and a candidate pruning strategy. © 2024 IEEE.

关键词： AI4DB Index Advisors Robustness Assessment

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：