检索结果-内蒙古大学图书馆

38th Conference on Neural Information Processing Systems, NeurIPS 2024

作者： Fu, Shenghao Yan, Junkai Yang, Qize Wei, Xihan Xie, Xiaohua Zheng, Wei-Shi School of Computer Science and Engineering Sun Yat-sen University China Peng Cheng Laboratory Shenzhen518055 China Tongyi Lab Alibaba Group China Key Laboratory of Machine Intelligence and Advanced Computing Ministry of Education China Guangdong Province Key Laboratory of Information Security Technology China Guangdong Guangzhou510555 China

Recent vision foundation models can extract universal representations and show impressive abilities in various tasks. However, their application on object detection is largely overlooked, especially without fine-tuning them. In this work, we show that frozen foundation models can be a versatile feature enhancer, even though they are not pre-trained for object detection. Specifically, we explore directly transferring the high-level image understanding of foundation models to detectors in the following two ways. First, the class token in foundation models provides an in-depth understanding of the complex scene, which facilitates decoding object queries in the detector's decoder by providing a compact context. Additionally, the patch tokens in foundation models can enrich the features in the detector's encoder by providing semantic details. Utilizing frozen foundation models as plug-and-play modules rather than the commonly used backbone can significantly enhance the detector's performance while preventing the problems caused by the architecture discrepancy between the detector's backbone and the foundation model. With such a novel paradigm, we boost the SOTA query-based detector DINO from 49.0% AP to 51.9% AP (+2.9% AP) and further to 53.8% AP (+4.8% AP) by integrating one or two foundation models respectively, on the COCO validation set after training for 12 epochs with R50 as the detector's backbone. © 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

Memory-Guided Contrastive and Triplet Separation for Weakly-Supervised Disease Detection

Memory-Guided Contrastive and Triplet Separation for Weakly-...

引用

2024 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2024

作者： She, Jinwen Li, Qiong Ma, Andy J. Sun Yat-sen University Cancer Center Guangzhou China Sun Yat-sen University School of Computer Science and Engineering Guangzhou China Ministry of Education Guangdong Province Key Laboratory of Information Security Technology Key Laboratory of Machine Intelligence and Advanced Computing China

ISBN: (纸本)9798350386226

Weakly-supervised disease detection has the great potential to alleviate the time-consuming and labor-intensive burden of manual annotations in instance level. While existing methods extract normality prototypes encoding normal patterns to improve the detection performance, they may fail to detect subtle anomalies without considering the diverse abnormal patterns. In this paper, we propose to recognize both normal and abnormal patterns for feature representation learning in disease detection based on a dual memory network. To learn discriminative memory banks and classifiers, dual memory loss is incorporated with a feature magnitude separation loss for model training. With the proposed contrastive feature separation loss and triplet feature separation loss, feature discriminability is further improved to obtain a more separable decision boundary. Additionally, we collect a large-scale CT dataset to evaluate lung tumor detection. Extensive experiments on the publicly available PANDA-MIL dataset and the collected LUNG-MIL dataset demonstrate the superiority of our proposed method compared to the state-of-the-art approaches for weakly-supervised disease detection. © 2024 IEEE.

关键词： Contrastive Learning

来源：评论

学校读者我要写书评

暂无评论

EB-SNN: An Ensemble Binary Spiking Neural Network for Visual Recognition 27th

EB-SNN: An Ensemble Binary Spiking Neural Network for Visua...

引用

27th International Conference on Pattern Recognition, ICPR 2024

作者： Li, Xinjie Tang, Jianxiong Lai, Jianhuang School of Computer Science and Engineering Sun Yat-sen University Guangzhou China Guangzhou510555 China Key Laboratory of Machine Intelligence and Advanced Computing Ministry of Education Guangzhou China

ISBN: (纸本)9783031781858

In recent years, spiking neural networks (SNNs) have gained significant attention in visual recognition tasks due to the low computational energy. However, most SNNs have a large number of parameters, which limits their use on resource-limited devices. In this paper, we propose an Ensemble Binary Spiking Neural Network (EB-SNN) for accurate and memory-friendly visual recognition. The EB-SNN is modeled by Ensemble Binary Weights (EBW) module, which integrates multiple binary weights for lightweight SNN modeling. Meanwhile, we propose Knowledge Alignment Strategy to ensure that the EB-SNN can approximate a well-trained SNN for good performance. Experimental results show that the EB-SNN can achieve accuracy of 95.39% on CIFAR10, using 9.3% memory of full-precision SNN. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Local binary pattern

来源：评论

学校读者我要写书评

暂无评论

Breaking Information Silos: Global Guided Task Prediction for Class-Incremental Learning 27th

Breaking Information Silos: Global Guided Task Prediction fo...

引用

27th International Conference on Pattern Recognition, ICPR 2024

作者： Hu, Chaoshun Ye, Biaohua Chen, Zixuan Lai, Jian-Huang School of Computer Science and Engineering Sun Yat-sen University Guangzhou China Guangzhou510555 China Ministry of Education Key Laboratory of Machine Intelligence and Advanced Computing Guangzhou China

ISBN: (纸本)9783031781889

Class-incremental learning (CIL) aims to learn a series of tasks sequentially, each introducing several new categories. Because providing the task labels during inference can significantly increase accuracy, many approaches attempt to predict task labels in CIL. However, existing works focus on learning local information and overlook the importance of global information. The absence of global guidance leads to the formation of information silos across disparate tasks, resulting in potential inter-task interference. To break information silos, we propose a method called Global Guided Task Prediction (GGTP) to introduce global information. Our method consists of two modules. The local de-redundant module aims to reduce information redundancy across different tasks from a global perspective. We combine dual encoders and feature decorrelation loss to effectively reduce redundancy while minimizing catastrophic forgetting. The global information module explicitly extracts global information to serve as auxiliary information guiding task prediction. We extract the most important information from all local information through a global encoder and then aggregate them to form global information. Extensive experiments validate the effectiveness of our method, which achieves state-of-the-art results. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Redundancy

来源：评论

学校读者我要写书评

暂无评论

TexCIL: Text-Guided Continual Learning of Disease with Vision-Language Model

TexCIL: Text-Guided Continual Learning of Disease with Visio...

引用

2024 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2024

作者： Zhang, Wentao Zhao, Defeng Zheng, Wei-Shi Wang, Ruixuan Sun Yat-sen University School of Computer Science and Engineering Guangzhou China Peng Cheng Laboratory Shenzhen China Moe Key Laboratory of Machine Intelligence and Advanced Computing Guangzhou China

ISBN: (纸本)9798350386226

Current intelligent diagnostic systems often catastrophically forget old knowledge when learning new diseases only from the training dataset of the new diseases. Inspired by human learning of visual classes with the effective help of language, we propose a continual learning framework based on a pre-trained visual-language model (VLM) without storing any image of previously learned diseases. In this framework, textual prior knowledge of each new disease can be obtained by utilizing the frozen VLM's text encoder, and then used to guide the visual learning of the new disease. This framework innovatively utilizes the textual prior knowledge of all previously learned diseases as out-of-distribution (OOD) information to help differentiate currently being-learned diseases from others. Extensive empirical evaluations on both medical and natural image datasets confirm the superiority of the proposed method over existing state-of-the-art methods in continual learning of new visual classes. The source code is available at https://***/OpenMedIA/TexCIL. © 2024 IEEE.

关键词： Contrastive Learning

来源：评论

学校读者我要写书评

暂无评论

Continual learning with Bayesian model based on a fixed pre-trained feature extractor

引用

Visual intelligence 2023年第1期1卷 1-14页

作者： Yang, Yang Cui, Zhiying Xu, Junjie Zhong, Changhong Zheng, Wei-Shi Wang, Ruixuan School of Computer Science and Engineering Sun Yat-sen University Guangzhou China Department of Network Intelligence Peng Cheng Laboratory Shenzhen China Key Laboratory of Machine Intelligence and Advanced Computing MOE Guangzhou China

Deep learning has shown its human-level performance in various applications. However, current deep learning models are characterized by catastrophic forgetting of old knowledge when learning new classes. This poses a challenge such as in intelligent diagnosis systems where initially only training data of a limited number of diseases are available. In this case, updating the intelligent system with data of new diseases would inevitably downgrade its performance on previously learned diseases. Inspired by the process of learning new knowledge in human brains, we propose a Bayesian generative model for continual learning built on a fixed pre-trained feature extractor. In this model, knowledge of each old class can be compactly represented by a collection of statistical distributions, e.g., with Gaussian mixture models, and naturally kept from forgetting in continual learning over time. Unlike existing class-incremental learning methods, the proposed approach is not sensitive to the continual learning process and can be additionally well applied to the data-incremental learning scenario. Experiments on multiple medical and natural image classification tasks reveal that the proposed approach outperforms state-of-the-art approaches that even keep some images of old classes during continual learning of new classes. © The Author(s) 2023.

关键词： Bayesian model Continual learning Fixed feature extractor Generative approach

来源：评论

学校读者我要写书评

暂无评论

Grasp as You Say: Language-guided Dexterous Grasp Generation 38

Grasp as You Say: Language-guided Dexterous Grasp Generation

引用

38th Conference on Neural Information Processing Systems, NeurIPS 2024

作者： Wei, Yi-Lin Jiang, Jian-Jian Xing, Chengyi Tan, Xian-Tuo Wu, Xiao-Ming Li, Hao Cutkosky, Mark Zheng, Wei-Shi School of Computer Science and Engineering Sun Yat-sen University China Stanford University United States Key Laboratory of Machine Intelligence and Advanced Computing Ministry of Education China

This paper explores a novel task "Dexterous Grasp as You Say" (DexGYS), enabling robots to perform dexterous grasping based on human commands expressed in natural language. However, the development of this field is hindered by the lack of datasets with natural human guidance;thus, we propose a language-guided dexterous grasp dataset, named DexGYSNet, offering high-quality dexterous grasp annotations along with flexible and fine-grained human language guidance. Our dataset construction is cost-efficient, with the carefully-design hand-object interaction retargeting strategy, and the LLM-assisted language guidance annotation system. Equipped with this dataset, we introduce the DexGYSGrasp framework for generating dexterous grasps based on human language instructions, with the capability of producing grasps that are intent-aligned, high quality and diversity. To achieve this capability, our framework decomposes the complex learning process into two manageable progressive objectives and introduce two components to realize them. The first component learns the grasp distribution focusing on intention alignment and generation diversity. And the second component refines the grasp quality while maintaining intention consistency. Extensive experiments are conducted on DexGYSNet and real world environments for validation. © 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

Exploiting Multiple Features for Hash Codes Learning with Semantic-Alignment-Promoting Variational Auto-encoder 12th

Exploiting Multiple Features for Hash Codes Learning with Se...

引用

12th National CCF Conference on Natural Language Processing and Chinese computing, NLPCC 2023

作者： Chen, Jiayang Su, Qinliang School of Computer Science and Engineering Sun Yat-Sen University Guangzhou China Guangdong Key Laboratory of Big Data Analysis and Processing Guangzhou China Key Laboratory of Machine Intelligence and Advanced Computing Ministry of Education Beijing China

ISBN: (纸本)9783031446924

Semantic hashing is an effective technique to empower information retrieval. Currently, considerable efforts have been dedicated to generating high-quality hash codes by modeling document features using generative models and other approaches. However, most of these methods rely solely on a single type of feature, such as TFIDF features, BERT embeddings, etc. As different types of features have distinct but complementary information of documents, e.g. TFIDF mainly contains the keywords information and BERT focuses on the semantics, hash codes generated solely from either may not capture the full essence of the documents. To overcome this challenge, we propose a semantic-alignment-promoting variational auto-encoder to generate hash codes from multiple document features. Specifically, a VAE-based generative model is first developed to model the multiple features. Then, we propose a semantic-alignment-promoting inference network to estimate the parameters of the variational posterior from multiple features. Additionally, the quality of hash codes is further improved by promoting the semantic alignment between the hash codes of connected documents in a constructed connection graph. The results of extensive experiments on three public datasets demonstrate that our proposed model significantly outperforms current state-of-the-art models. © 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.

关键词： Semantics

来源：评论

学校读者我要写书评

暂无评论

Organic-inorganic hybrid hole transport layers with SnS doping boost the performance of perovskite solar cells

引用

Journal of Energy Chemistry 2022年第5期31卷 637-645页

作者： Xiaolu Zheng Haibing Wang Feihong Ye Cong Chen Weijun Ke Wenjing Zhang Chuanxin He Yanlong Tai Guojia Fang College of Chemistry and Environmental Engineering Institute of Microscale OptoelectronicsShenzhen UniversityShenzhen 518060GuangdongChina Shenzhen Institute Wuhan UniversityShenzhen 518057GuangdongChina Key Laboratory of Human-Machine Intelligence-Synergy Systems of Chinese Academy of Sciences and Guangdong-Hong Kong-Macao Joint Laboratory of Human-Machine Intelligence-Synergy Systems Shenzhen Institutes of Advanced TechnologyChinese Academy of SciencesShenzhen 518055GuangdongChina

Perovskite solar cells(PSCs) have demonstrated excellent photovoltaic performance which currently rival the long-standing silicon solar cells’ efficiency. However, the relatively poor device operational stability of PSCs still limits their future commercialization. Binary sulfide is a category of materials with promising optoelectrical properties, which shows the potential to improve both the efficiency and stability of *** we demonstrate that the inorganic tin monosulfide(Sn S) can be an efficient dopant in 2,2’,7,7’-tet rakis(N,N-di-p-methoxy-phenylamine)-9,9’-spirobifluorene(spiro-OMe TAD) to form a composite hole transport layer(HTL) for PSCs. Sn S nanoparticles(NPs) synthesized through a simple chemical precipitation method exhibit good crystallization and suitable band matching with the perovskites. The introduction of Sn S NPs in Spiro-OMTAD HTLs enhanced charge extraction, reduced trap state density, and shallowed trap state energy level of the devices based on the composite HTLs. Therefore, the resulting solar cells employing Sn S-doped spiro-OMe TAD HTLs delivered an improved stabilized power output efficiency of 21.75% as well as enhanced long-term stability and operational stability. Our results provide a simple method to modify the conventional spiro-OMe TAD and obtain PSCs with both high efficiency and good stability.

关键词： Perovskite solar cells SnS nanoparticles Trap state passivation Stability

来源：评论

学校读者我要写书评

暂无评论

Imputing DNA Methylation by Transferred Learning Based Neural Network

引用

Journal of Computer Science & Technology 2022年第2期37卷 320-329页

作者： Xin-Feng Wang Xiang Zhou Jia-Hua Rao Zhu-Jin Zhang Yue-Dong Yang School of Computer Science and Engineering Sun Yat-sen UniversityGuangzhou 510000China Key Laboratory of Machine Intelligence and Advanced Computing of Ministry of Education(Sun Yat-sen University)Guangzhou 510000 China

DNA methylation is one important epigenetic type to play a vital role in many diseases including *** the development of the high-throughput sequencing technology,there is much progress to disclose the relations of DNA methylation with ***,the analyses of DNA methylation data are challenging due to the missing values caused by the limitations of current *** many methods have been developed to impute the missing values,these methods are mostly based on the correlations between individual samples,and thus are limited for the abnormal samples in *** this study,we present a novel transfer learning based neural network to impute missing DNA methylation data,namely the TDimpute-DNAmeth *** method learns common relations between DNA methylation from pan-cancer samples,and then fine-tunes the learned relations over each specific cancer type for imputing the missing *** on 16 cancer datasets,our method was shown to outperform other commonly-used *** analyses indicated that DNA methylation is related to cancer survival and thus can be used as a biomarker of cancer prognosis.

关键词： neural network transfer learning DNA methylation data imputation survival analysis

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：