检索结果-内蒙古大学图书馆

From Missing Pieces to Masterpieces: Image Completion with Context-Adaptive Diffusion

IEEE Transactions on Pattern Analysis and Machine Intelligence 2025年第7期47卷 6073-6087页

作者： Shamsolmoali, Pourya Zareapoor, Masoumeh Zhou, Huiyu Felsberg, Michael Tao, Dacheng Li, Xuelong University of York Department of Computer Science United Kingdom University of Leicester School of Computing and Mathematical Sciences United Kingdom Linkoping University Computer Vision Laboratory Sweden Nanyang Technological University College of Computing & Data Science Singapore China Northwestern Polytechnical University Key Laboratory of Intelligent Interaction and Applications Ministry of Industry and Information Technology Xi'an China

Image completion is a challenging task, particularly when ensuring that generated content seamlessly integrates with existing parts of an image. While recent diffusion models have shown promise, they often struggle with maintaining coherence between known and unknown (missing) regions. This issue arises from the lack of explicit spatial and semantic alignment during the diffusion process, resulting in content that does not smoothly integrate with the original image. Additionally, diffusion models typically rely on global learned distributions rather than localized features, leading to inconsistencies between the generated and existing image parts. In this work, we propose ConFill, a novel framework that introduces a Context-Adaptive Discrepancy (CAD) model to ensure that intermediate distributions of known and unknown regions are closely aligned throughout the diffusion process. By incorporating CAD, our model progressively reduces discrepancies between generated and original images at each diffusion step, leading to contextually aligned completion. Moreover, ConFill uses a new Dynamic Sampling mechanism that adaptively increases the sampling rate in regions with high reconstruction complexity. This approach enables precise adjustments, enhancing detail and integration in restored areas. Extensive experiments demonstrate that ConFill outperforms current methods, setting a new benchmark in image completion. © 1979-2012 IEEE.

关键词： computer aided design

来源：评论

学校读者我要写书评

暂无评论

Sampling Methods for Efficient Training of Graph Convolutional Networks:A Survey

引用

IEEE/CAA Journal of Automatica Sinica 2022年第2期9卷 205-234页

作者： Xin Liu Mingyu Yan Lei Deng Guoqi Li Xiaochun Ye Dongrui Fan State Key Laboratory of Computer Architecture Institute of Computing TechnologyChinese Academy of SciencesBeijing 100086 School of Computer Science and Technology University of Chinese Academy of SciencesBeijing 100049 IEEE Department of Precision Instrument Center for Brain Inspired Computing ResearchTsinghua UniversityBeijing 100084China

Graph convolutional networks(GCNs)have received significant attention from various research fields due to the excellent performance in learning graph *** GCN performs well compared with other methods,it still faces *** a GCN model for large-scale graphs in a conventional way requires high computation and storage ***,motivated by an urgent need in terms of efficiency and scalability in training GCN,sampling methods have been proposed and achieved a significant *** this paper,we categorize sampling methods based on the sampling mechanisms and provide a comprehensive survey of sampling methods for efficient training of *** highlight the characteristics and differences of sampling methods,we present a detailed comparison within each category and further give an overall comparative analysis for the sampling methods in all ***,we discuss some challenges and future research directions of the sampling methods.

关键词： Efficient training graph convolutional networks(GCNs) graph neural networks(GNNs) sampling method

来源：评论

学校读者我要写书评

暂无评论

Bubble transfer on wettability-heterogeneous surfaces

引用

Chinese Chemical Letters 2023年第7期34卷 472-476页

作者： Chunhui Zhang Xiao Xiao Ziwei Guo Lei Jiang Cunming Yu CAS Key Laboratory of Bio-inspired Materials and Interfacial Science Technical Institute of Physics and ChemistryBeijing 100190China School of Chemical Sciences University of Chinese Academy of SciencesBeijing 100049China Key Laboratory of Bio-inspired Smart Interfacial Science and Technology of Ministry of Education School of ChemistryBeihang UniversityBeijing 100191China Department of Electrical and Computer Engineering National University of SingaporeSingapore 117583Singapore

Researches have investigated the formation, transportation and spreading of bubble on solid surface with specific wettability. However, bubble transfer on wettability-heterogeneous surfaces has been rarely reported, which also plays significant role in water electrolysis, heat transfer, micro-bubble collection, *** this work, we carefully investigate the behavior of bubble transfer from the aerophobic or aerophilic region to the superaerophilic region through fabricating the wettability-heterogenous surfaces. Surface energy was elucidated to be transformed to the kinetic energy during bubble transfer process. Theoretical analysis on the average velocity of bubble transfer was consistent with the experimental *** influence of wettability of solid substrate, bubble volume and superaerophilic stripe width on bubble transfer are carefully investigated. Moreover, wettability-heterogeneous surfaces were explored to be applied in micro-CO_(2) bubble collection and H_(2) bubble removement in water splitting.

关键词： Bubble transfer Superaerophilicity Wettability-heterogenous Surface energy Bubble collection Water splitting

来源：评论

学校读者我要写书评

暂无评论

Learning-Based Problem Reduction for Large-Scale Uncapacitated Facility Location Problems 13

Learning-Based Problem Reduction for Large-Scale Uncapacitat...

引用

13th IEEE Congress on Evolutionary Computation, CEC 2024

作者： Zhang, Shuaixiang Yang, Yixuan Tong, Hao Yao, Xin Lingnan University Department of Computing and Decision Sciences Hong Kong Southern University of Science and Technology Guangdong Provincial Key Laboratory of Brain-inspired Intelligent Computation Department of Computer Science and Engineering Shenzhen China

ISBN: (纸本)9798350308365

The uncapacitated facility location problem (UFLP) is a well-known combinational optimization problem, attracting numerous heuristic and meta-heuristic methods. However, these effective algorithms still encounter challenges when solving large-scale UFLP instances. To deal with the thousands of decision variables, one possible approach is to predict and remove unimportant facilities and thus decreases the dimensionality of problem instances. This paper explores the problem reduction of large-scale instances of UFLP. We propose suitable features of facilities to construct machine learning models for UFLP by learning from small instances. Based on this model, we introduce a novel problem reduction strategy-based optimization framework, and apply it to large instances. Through comprehen-sive experiments, we show that the proposed problem reduction strategy can effectively transform the UFLP instances to smaller ones. The performance of existing solution methods can be significantly enhanced, especially on large-scale instances. Under the proposed framework, we also show the generalization abilities of our models, which can be improved further. © 2024 IEEE.

关键词： Adversarial machine learning

来源：评论

学校读者我要写书评

暂无评论

Evolutionary Reinforcement Learning via Cooperative Coevolution 27

Evolutionary Reinforcement Learning via Cooperative Coevolut...

引用

27th European Conference on Artificial Intelligence, ECAI 2024

作者： Hu, Chengpeng Liu, Jialin Yao, Xin Guangdong Provincial Key Laboratory of Brain-inspired Intelligent Computation Department of Computer Science and Engineering Southern University of Science and Technology Shenzhen China Research Institute of Trustworthy Autonomous System Southern University of Science and Technology Shenzhen China Department of Computing and Decision Sciences Lingnan University Hong Kong

ISBN: (纸本)9781643685489

Recently, evolutionary reinforcement learning has obtained much attention in various *** a population of actors, evolutionary reinforcement learning utilises the collected experiences to improve the behaviour policy through efficient ***, the poor scalability of genetic operators limits the efficiency of optimising high-dimensional neural *** address this issue, this paper proposes a novel cooperative coevolutionary reinforcement learning (CoERL) *** by cooperative coevolution, CoERL periodically and adaptively decomposes the policy optimisation problem into multiple subproblems and evolves a population of neural networks for each of the *** of using genetic operators, CoERL directly searches for partial gradients to update the *** policy with partial gradients maintains consistency between the behaviour spaces of parents and offspring across *** experiences collected by the population are then used to improve the entire policy, which enhances the sampling *** on six benchmark locomotion tasks demonstrate that CoERL outperforms seven state-of-the-art algorithms and *** study verifies the unique contribution of CoERL’s core ingredients. © 2024 The Authors.

关键词： Adversarial machine learning

来源：评论

学校读者我要写书评

暂无评论

Reliability-Aware Network Slicing based on Multi-Objective Optimization 13

Reliability-Aware Network Slicing based on Multi-Objective O...

引用

13th IEEE Congress on Evolutionary Computation, CEC 2024

作者： Xia, Qiqi Wang, Jin Huang, Chengqiang Yao, Xin Southern University of Science and Technology Shenzhen China Southern University of Science and Technology Guangdong Key Laboratory of Brain-inspired Intelligent Computation Department of Computer Science and Engineering Shenzhen China Huawei Technologies Co. Ltd. Reliability Technology Lab Shenzhen China Lingnan University Department of Computing and Decision Sciences Hong Kong

ISBN: (纸本)9798350308365

Network slicing is an essential technology in 5G and the forthcoming 6G networks. It aims to embed multiple virtual networks, i.e., network slices, on top of a shared substrate network to meet diverse service requirements. While a considerable body of existing research strives to maximize overall profits by meeting the resource demands of the network slices, optimizing their reliability is frequently overlooked. In this paper, we formalize the network slicing problem as a multi-objective optimization problem that aims to maximize total profits and reliability of network slices. To tackle this problem, we propose a new multi-objective optimization approach that improves over the state-of-the-art algorithm, which can achieve good approximate Pareto front results balancing total profits and reliability of network slices. The performance of our proposed method is evaluated on both artificial and real-world network topologies. Experimental results demonstrate the superior performance of our proposed method compared to the baseline algorithm, outperforming the latter in 92% of instances in terms of the Hypervolume (HV) metric. © 2024 IEEE.

关键词： Optimization algorithms

来源：评论

学校读者我要写书评

暂无评论

A Study of Fitness Gains in Evolving Finite State Machines

arXiv

引用

arXiv 2023年

作者： Zoltai, Gábor Xie, Yue Neumann, Frank Optimisation and Logistics School of Computer and Mathematical Sciences University of Adelaide Australia Bio-Inspired Robotics Laboratory Department of Engineering University of Cambridge United Kingdom

Among the wide variety of evolutionary computing models, Finite State Machines (FSMs) have several attractions for fundamental research. They are easy to understand in concept and can be visualised clearly in simple cases. They have a ready fitness criterion through their relationship with Regular Languages. They have also been shown to be tractably evolvable, even up to exhibiting evidence of open-ended evolution in specific scenarios. In addition to theoretical attraction, they also have industrial applications, as a paradigm of both automated and user-initiated control. Improving the understanding of the factors affecting FSM evolution has relevance to both computer science and practical optimisation of control. We investigate an evolutionary scenario of FSMs adapting to recognise one of a family of Regular Languages by categorising positive and negative samples, while also being under a counteracting selection pressure that favours fewer states. The results appear to indicate that longer strings provided as samples reduce the speed of fitness gain, when fitness is measured against a fixed number of sample strings. We draw the inference that additional information from longer strings is not sufficient to compensate for sparser coverage of the combinatorial space of positive and negative sample strings. © 2023, CC BY-NC-ND.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Procedural Fairness in Machine Learning

arXiv

引用

arXiv 2024年

作者： Wang, Ziming Huang, Changwu Yao, Xin Shenzhen518055 China Guangdong Provincial Key Laboratory of Brain-inspired Intelligent Computation Department of Computer Science and Engineering of SUSTech China Department of Computing and Decision Sciences Lingnan University Hong Kong School of Computer Science University of Birmingham United Kingdom

Fairness in machine learning (ML) has received much attention. However, existing studies have mainly focused on the distributive fairness of ML models. The other dimension of fairness, i.e., procedural fairness, has been neglected. In this paper, we first define the procedural fairness of ML models, and then give formal definitions of individual and group procedural fairness. We propose a novel metric to evaluate the group procedural fairness of ML models, called GPFFAE, which utilizes a widely used explainable artificial intelligence technique, namely feature attribution explanation (FAE), to capture the decision process of the ML models. We validate the effectiveness of GPFFAE on a synthetic dataset and eight real-world datasets. Our experiments reveal the relationship between procedural and distributive fairness of the ML model. Based on our analysis, we propose a method for identifying the features that lead to the procedural unfairness of the model and propose two methods to improve procedural fairness after identifying unfair features. Our experimental results demonstrate that we can accurately identify the features that lead to procedural unfairness in the ML model, and both of our proposed methods can significantly improve procedural fairness with a slight impact on model performance, while also improving distributive fairness. Copyright © 2024, The Authors. All rights reserved.

关键词： Machine learning

来源：评论

学校读者我要写书评

暂无评论

A Knowledge Guided Multi-Population Evolutionary Algorithm for Dynamic Workflow Scheduling Problem

A Knowledge Guided Multi-Population Evolutionary Algorithm f...

引用

Artificial Intelligence (CAI), IEEE Conference on

作者： Jingyuan Xu Jiajian Yang Peiru Li Ziming Wang Changwu Huang Xin Yao Department of Computer Science and Engineering Guangdong Provincial Key Laboratory of Brain-Inspired Intelligent Computation Southern University of Science and Technology Shenzhen China Department of Computing and Decision Sciences Lingnan University Hong Kong SAR

ISBN: (数字)9798350354096

ISBN: (纸本)9798350354102

The workflow scheduling problem is a fundamental task in cloud computing. This paper addresses the challenge of workflow scheduling in dynamic and uncertain cloud environments, where computing resources may become inaccessible due to hardware or software failures. To tackle this challenge, we propose a novel algorithm called the Order Feature Guided Multi-Population (OFGMP) algorithm for dynamic workflow scheduling in cloud environments. The OFGMP algorithm utilizes a multi-population evolutionary framework, incorporating a knowledge-guided reproduction operator that leverages the order feature of solutions, as well as repair mechanisms to adapt to changing environmental conditions. Extensive experiments are conducted to validate the algorithm’s performance against existing dynamic scheduling approaches. The experimental results demonstrate the superiority of our proposed method over others on a number of test cases.

关键词： Cloud computing Processor scheduling Heuristic algorithms Software algorithms Evolutionary computation Maintenance engineering Dynamic scheduling

来源：评论

学校读者我要写书评

暂无评论

Learning-Based Problem Reduction for Large-Scale Uncapacitated Facility Location Problems

Learning-Based Problem Reduction for Large-Scale Uncapacitat...

引用

Congress on Evolutionary Computation

作者： Shuaixiang Zhang Yixuan Yang Hao Tong Xin Yao Department of Computer Science and Engineering Guangdong Provincial Key Laboratory of Brain-inspired Intelligent Computation Southern University of Science and Technology Shenzhen China Department of Computing and Decision Sciences Lingnan University Hong Kong SAR China

ISBN: (数字)9798350308365

ISBN: (纸本)9798350308372

关键词： Dimensionality reduction Metaheuristics Training data Transforms Evolutionary computation Benchmark testing Prediction algorithms

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：