检索结果-内蒙古大学图书馆

Leveraging differentiable NAS and abstract genetic algorithms for optimizing on-mobile VSR performance

Machine Learning 2025年第7期114卷 1-23页

作者： Liu, Xuncheng Zhang, Weizhan Gong, Tieliang Yan, Caixia Li, Rui School of Computer Science and Technology Xi’an Jiaotong University Xi’an China Ministry of Education Key Laboratory of Intelligent Networks and Network Security Xi’an Jiaotong University Xi’an China Shaanxi Province Key Laboratory of Big Data Knowledge Engineering Xi’an Jiaotong University Xi’an China School of Continuing Education Xi’an Jiaotong University Xi’an China

The proliferation of mobile video applications has for real-time Video Super-Resolution (VSR) technologies, necessitating architectures that simultaneously achieve superior fidelity and ultra-low latency. Existing Neural Architecture Search (NAS) methods achieve efficient performance within fixed topologies and feature dimensionalities, but inherently struggle to meet the flexible design demands of lightweight on-mobile architectures. When applied to on-mobile VSR, their insufficient exploration under real-time constraints results in suboptimal performance, exposing a critical gap in current approaches. To address these challenges, we propose a novel bilevel NAS framework designed to systematically explore diverse topological configurations and feature dimensionalities within an expanded search space. Our framework incorporates a collection of independent differentiable search spaces. At the upper level, we employ an Abstract Genetic Algorithm (AGA) to explore a broad spectrum of topological variations and feature dimensionality configurations. At the lower level, we introduce a latency-aware differentiable NAS method, which facilitates efficient fidelity-latency optimization within each differentiable search space. Additionally, we provide theoretical convergence guarantees for our bilevel NAS framework, ensuring the discovery of optimal architectures. Experimental validation demonstrates that our method achieves state-of-the-art performance, delivering a 5.78ms inference latency while significantly improving fidelity metrics across standard benchmarks. The proposed framework not only breaks the fidelity-latency trade-off but also demonstrates remarkable efficiency in early AGA generations, reducing architecture search iterations from 100 to several generations while maintaining superior results. Our implementation is available at https://***/liuxunchenglxc/Reducio .

关键词：

来源：评论

学校读者我要写书评

暂无评论

SwimVG: Step-wise Multimodal Fusion and Adaption for Visual Grounding

arXiv

引用

arXiv 2025年

作者： Shi, Liangtao Liu, Ting Hu, Xiantao Hu, Yue Yin, Quanjun Hong, Richang Key Laboratory of Knowledge Engineering with Big Data Hefei University of Technology Hefei230009 China Ministry of Education and School of Computer Science and Information Engineering Hefei University of Technology Hefei230009 China School of systems engineering National University of Defense Technology Hunan Province Changsha410073 China Department of Computer Science and Engineering Nanjing University of Science and Technology Nanjing210014 China

Visual grounding aims to ground an image region through natural language, which heavily relies on cross-modal alignment. Most existing methods transfer visual/linguistic knowledge separately by fully fine-tuning uni-modal pre-trained models, followed by a simple stack of visual-language transformers for multimodal fusion. However, these approaches not only limit adequate interaction between visual and linguistic contexts, but also incur significant computational costs. Therefore, to address these issues, we explore a step-wise multimodal fusion and adaption framework, namely SwimVG. Specifically, SwimVG proposes step-wise multimodal prompts (Swip) and cross-modal interactive adapters (CIA) for visual grounding, replacing the cumbersome transformer stacks for multimodal fusion. Swip can improve the alignment between the vision and language representations step by step, in a token-level fusion manner. In addition, weight-level CIA further promotes multimodal fusion by cross-modal interaction. Swip and CIA are both parameter-efficient paradigms, and they fuse the cross-modal features from shallow to deep layers gradually. Experimental results on four widely-used benchmarks demonstrate that SwimVG achieves remarkable abilities and considerable benefits in terms of efficiency. Our code is available at https://***/liuting20/SwimVG. Copyright © 2025, The Authors. All rights reserved.

关键词： Visual languages

来源：评论

学校读者我要写书评

暂无评论

A Zero-shot Learning Method with a Multi-Modal knowledge Graph

A Zero-shot Learning Method with a Multi-Modal Knowledge Gra...

引用

International Conference on Tools for Artificial Intelligence (ICTAI)

作者： Yuhong Zhang Haitao Shu Chenyang Bu Xuegang Hu Key Laboratory of Knowledge Engineering with Big Data Ministry of Education Hefei China School of Computer Science and Information Engineering Hefei University of Technology Hefei China

Zero-shot learning aims to recognize unseen-classes using some seen-class samples as training set. It is challenging owing to that the feature representations of unseen-class samples are unavailable. Existing methods transfer the mapping from seen-classes to unseen-classes with the correlation as a bridge, in which, the semantic representations are used to discriminate the classes. However, the unavailability of visual representations for unseen-classes and the insufficient discrimination of semantic representations make the zero-shot learning challenging. Therefore, the visual representations are learned as complements to semantic representations to construct a multi-modal knowledge graph (KG), and a zero-shot learning method based on multi-modal KG is proposed in this paper. Specially, a semantic KG is introduced to capture the correlation of classes, and with the correlation, the visual feature representations of all classes are learned. Then, the discriminative visual representations and the semantic representations are used together to construct a multi-modal KG. With the multi-modal KG, the classifier for seen-classes is transferred to unseen classes. Extensive experimental results show the effectiveness of our method.

关键词： Learning systems Representation learning Training Bridges Visualization Correlation Semantics

来源：评论

学校读者我要写书评

暂无评论

A Review-aware Graph Contrastive Learning Framework for Recommendation

arXiv

引用

arXiv 2022年

作者： Shuai, Jie Sun, Peijie Zhang, Kun Hong, Richang Li, Yong Wu, Le Wang, Meng Key Laboratory of Knowledge Engineering with Big Data Hefei University of Technology China Key Laboratory of Knowledge Engineering with Big Data Hefei University of Technology Institute of Artificial Intelligence Hefei Comprehensive National Science Center China Beijing National Research Center for Information Science and Technology Department of Electronic Engineering Tsinghua University China

Most modern recommender systems predict users’ preferences with two components: user and item embedding learning, followed by the user-item interaction modeling. By utilizing the auxiliary review information accompanied with user ratings, many of the existing review-based recommendation models enriched user/item embedding learning ability with historical reviews or better modeled user-item interactions with the help of available user-item target reviews. Though significant progress has been made, we argue that current solutions for review-based recommendation suffer from two drawbacks. First, as review-based recommendation can be naturally formed as a user-item bipartite graph with edge features from corresponding user-item reviews, how to better exploit this unique graph structure for recommendation? Second, while most current models suffer from limited user behaviors, can we exploit the unique self-supervised signals in the review-aware graph to guide two recommendation components better? To this end, in this paper, we propose a novel Review-aware Graph Contrastive Learning (RGCL) framework for review-based recommendation. Specifically, we first construct a review-aware user-item graph with feature-enhanced edges from reviews, where each edge feature is composed of both the user-item rating and the corresponding review semantics. This graph with feature-enhanced edges can help attentively learn each neighbor node weight for user and item representation learning. After that, we design two additional contrastive learning tasks (i.e., Node Discrimination and Edge Discrimination) to provide self-supervised signals for the two components in recommendation process. Finally, extensive experiments over five benchmark datasets demonstrate the superiority of our proposed RGCL compared to the state-of-the-art baselines. Copyright © 2022, The Authors. All rights reserved.

关键词： Recommender systems

来源：评论

学校读者我要写书评

暂无评论

Multi-lingual acquisition on multimodal pre-training for cross-modal retrieval 22

Multi-lingual acquisition on multimodal pre-training for cro...

引用

Proceedings of the 36th International Conference on Neural Information Processing Systems

作者： Liang Zhang Anwen Hu Qin Jin School of Information Renmin University of China School of Information Renmin University of China and Key Laboratory of Data Engineering and Knowledge Engineering (MOE) Renmin University of China

ISBN: (纸本)9781713871088

Vision and diverse languages are important information sources in our living world. A model that understands multi-modalities and multi-languages can be applied to a wider range of real-life scenarios. To build such a multimodal and multilingual model, existing works try to ensemble vision-language data from multiple languages in pre-training. However, due to the large number of languages, these works often require huge computing resources and cannot be flexibly extended to new languages. In this work, we propose a Multi-Lingual Acquisition (MLA) framework that can easily empower a monolingual Vision-Language Pre-training (VLP) model with multilingual capability. Specifically, we design a lightweight language acquisition encoder based on state-of-the-art monolingual VLP models. We further propose a two-stage training strategy to optimize the language acquisition encoder, namely the Native Language Transfer stage and the Language Exposure stage. With much less multilingual training data and computing resources, our model achieves state-of-the-art performance on multilingual image-text and video-text retrieval benchmarks.

关键词：

来源：评论

学校读者我要写书评

暂无评论

引用

International Conference on Tools for Artificial Intelligence (ICTAI)

作者： Xinzheng Li Yuhong Zhang Peipei Li Xuegang Hu Key Laboratory of Knowledge Engineering with Big Data Ministry of Education Hefei China School of Computer Science and Information Engineering Hefei University of Technology Hefei China

Cross-network node classification aims to use a labeled source network to classify nodes in an unlabeled target network. Most of the existing cross-network node classification methods learn the network representations by capturing the node neighborhood and train the classifier on these representations. The performance is highly dependent on the high-quality neighborhood in the network. However, in applications, the degree of nodes generally follows a long-tail distribution, i.e., a significant proportion of nodes are tail nodes with sparse neighborhood. It poses a challenge to existing methods. To this end, a structure similarity graph for cross-network node classitication method (SCNC) is proposed in this paper. Firstly, the potential links between nodes are predicted with the structural similarity metric to construct structure similarity graph, which can enrich the neighborhood of tail nodes. Then, the embedding representations of the structural similarity graph are learned to capture more neighborhood information. Finally, the adversarial is used to learn the domain invariant representations to address cross-network divergence. Extensive experimental results show that our SCNC outperforms the state-of-the-art methods.

关键词： Network topology Tail Learning (artificial intelligence) Prediction algorithms Classification algorithms Noise measurement Task analysis

来源：评论

学校读者我要写书评

暂无评论

KAN v.s. MLP for Offline Reinforcement Learning

arXiv

引用

arXiv 2024年

作者： Guo, Haihong Li, Fengxin Li, Jiao Liu, Hongyan School of Information Renmin University of China China Institute of Medical Information Medical Library Chinese Academy of Medical Sciences Peking Union Medical College China Key Laboratory of Data Engineering and Knowledge Engineering Ministry of Education China School of Economics and Management Tsinghua University China

Kolmogorov-Arnold Networks (KAN) is an emerging neural network architecture in machine learning. It has greatly interested the research community about whether KAN can be a promising alternative of the commonly used Multi-Layer Perceptions (MLP). Experiments in various fields demonstrated that KAN-based machine learning can achieve comparable if not better performance than MLP-based methods, but with much smaller parameter scales and are more explainable. In this paper, we explore the incorporation of KAN into the actor and critic networks for offline reinforcement learning (RL). We evaluated the performance, parameter scales, and training efficiency of various KAN and MLP based conservative Q-learning (CQL) on the the classical D4RL benchmark for offline RL. Our study demonstrates that KAN can achieve performance close to the commonly used MLP with significantly fewer parameters. This provides us an option to choose the base networks according to the requirements of the offline RL tasks. © 2024, CC BY.

关键词： Adversarial machine learning

来源：评论

学校读者我要写书评

暂无评论

Mitigating Label-Noise for Facial Expression Recognition in the Wild

Mitigating Label-Noise for Facial Expression Recognition in ...

引用

2022 IEEE International Conference on Multimedia and Expo, ICME 2022

作者： Yan, Huan Gu, Yu Zhang, Xiang Wang, Yantong Ji, Yusheng Ren, Fuji School of Computer and Information Hefei University of Technology S2AC Lab China Hefei University of Technology Key Laboratory of Knowledge Engineering with Big Data China Anhui Province Key Laboratory of Cyberspace Security Situation Awareness and Evaluation China National Institute of Informatics Tokushima University Japan

ISBN: (数字)9781665485630

ISBN: (纸本)9781665485630

Label-noise constitutes a major challenge for facial expression recognition in the wild due to the ambiguity of facial expressions worsened by low-quality images. To deal with this problem, we propose a simple but effective Label-noise Robust Network (LRN) which explores the inter-class correlations for mitigating ambiguity that usually happens between morphologically similar classes. Specifically, LRN leverages a multivariate normal distribution to model such correlations at the final hidden layer of the neural network to suppress the heteroscedastic uncertainty caused by inter-class label noise. Furthermore, LRN utilizes a confidence-based label-free loss to extract compact intra-class feature representations under label noise while preserving the intrinsic inter-class relationships. Experiments on three in-the-wild facial expression datasets demonstrates the superiority of our method. © 2022 IEEE.

关键词： Face recognition

来源：评论

学校读者我要写书评

暂无评论

Automatic Retinal Nerve Fiber Trajectory Simulation and Quasi-polar Transformation for Detecting Retinal Nerve Fiber Layer Defect in Fundus Images

Automatic Retinal Nerve Fiber Trajectory Simulation and Quas...

引用

IEEE International Conference on Multimedia and Expo (ICME)

作者： Yanni Wang Gang Yang Dayong Ding Jianchun Zhao School of Information Renmin University of China Beijing China Key Lab of Data Engineering and Knowledge Engineering Renmin University of China Beijing China Vistel AI Lab Beijing Office Visionary Intelligence Ltd. Beijing China

The retinal nerve fiber layer defects (RNFLD) provide early objective evidence for many retinal abnormalities, especially early glaucoma. The recent success of deep learning has led to exciting prospects in automating the detection of RNFLD, but it is highly dependent on the availability of large-scale datasets carefully annotated by experienced ophthalmologists, which is both time-consuming and cost-prohibitive. Most previous works on RNFLD detection lack a unified annotation scheme. In addition, they fail to exploit the intrinsic morphological characteristics of retinal nerve fiber bundles (RNFB). This work presents an automatic RNFB tracing method that is applicable not only in automating the RNFLD annotation process, but also in performing a quasi-polar transformation on fundus images for subsequent RNFLD detection tasks. Also, our proposed method paves the way for alleviating the problem of inter-annotator variation incurred merely by different annotation habits among annotators and reducing the noise before the input stage. Experiments reveal that our proposed quasi-polar transformation provides a significant boost to the existing RNFLD detection method and surpasses the state-of-the-art F1 by 4.2%, and that our method also offers a more accurate and reasonable description of RNFLD detection results, which demonstrates the orientation of RNFLD rather than vague position indication.

关键词：

来源：评论

学校读者我要写书评

暂无评论

NcoVis: A Visual Analysis Framework for Exploring Academic Collaboration Networks under New Collaborative Relationships 25

NcoVis: A Visual Analysis Framework for Exploring Academic C...

引用

25th IEEE International Conference on Computer Supported Cooperative Work in Design, CSCWD 2022

作者： Yu, Ye Wu, Yao Liang, Xi Ma, Cheng Lu, Qiang Key Laboratory of Knowledge Engineering with Big Data Hefei University of Technology Ministry of Education Hefei China School of Computer Science and Information Hefei University of Technology Hefei China Intelligent Interconnected Systems Laboratory of Anhui Province Hefei University of Technology Hefei China

ISBN: (纸本)9781665405270

Academic collaboration has become one of the topics of large-scale social network analysis. The information extracted from academic collaboration also shows the contextual value in combination with other data. For academic collaborative social networks with intricate collaborative relationships, previous research methods mainly focused on network structure and community discovery, lacking in mining the characteristics of relationships in collaborative networks. There were some limitations in the research perspective. In this work, we propose a collaborative relationship representation method that is more suitable for academic collaborative social networks. It marks the attribute characteristics of nodes in collaborative relationships and highlights collaborative relationships in visualization. By adopting this new representation of collaborative relationships, we designed a visual analysis framework NcoVis, to visualize academic collaborative social networks. We further visually analyze the link relationship groups guided by nodes, and explore the potential representation of the structure and attribute characteristics of link relationships, so that analysts can explore the theme and structural change characteristics of link relationships, communities and the whole network in collaborative social networks. To prove the proposed framework, we conducted a case study on the academic collaboration literature data set, and analyzed the questionnaire with Likert scale through the user evaluation experiment, which verified the effectiveness of our method and the practicability of NcoVis. © 2022 IEEE.

关键词： Visualization

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：