检索结果-内蒙古大学图书馆

A Novel Ensemble Learning Method for Online Learning Scenarios 4

学校读者我要写书评

暂无评论

A Novel Ensemble Learning Method for Online Learning Scenari...

4th IEEE International Conference on Electronics technology, ICET 2021

作者： Liu, Ye Cao, Weipeng Liu, Yiwen Zou, Weidong Shenzhen University College of Computer Science and Software Engineering Shenzhen China Shenzhen University College of International Exchange Shenzhen China Beijing Institute of Technology School of Automation Beijing China

ISBN: (纸本)9781728176734

Online Sequential Extreme Learning Machine (OS-ELM) is a modeling algorithm for data stream mining that has attracted much attention in recent years. OS-ELM has extremely fast training speed and good generalization ability, however, because it uses random input weights for the model initialization and these parameters remain unchanged throughout the training process, which causes instability in the performance of the model. To alleviate this problem, we propose an improved OS-ELM algorithm based on the ensemble learning mechanism, which we call ELOS-ELM. ELOS-ELM uses multiple initialization methods to construct diverse base online learners, and the prediction results of the model are determined by voting by these base models. In this way, the final model can avoid misjudgments caused by the knowledge blind zone of a single base model. Extensive experimental results on eight benchmarks show that ELOS-ELM can effectively improve the stability and generalization ability of the OS-ELM. © 2021 IEEE.

关键词： E-learning

Look Inside for More: Internal Spatial Modality Perception for 3D Anomaly Detection

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Liang, Hanzhe Xie, Guoyang Hou, Chengbin Wang, Bingshu Gao, Can Wang, Jinbao College of Computer Science and Software Engineering Shenzhen University Shenzhen China Shenzhen Audencia Financial Technology Institute Shenzhen University Shenzhen China National Engineering Laboratory for Big Data System Computing Technology Shenzhen University Shenzhen China Department of Intelligent Manufacturing CATL Ningde China School of Computing and Artificial Intelligence Fuyao University of Science and Technology Fuzhou China School of Software Northwestern Polytechnical University Xi’an China Guangdong Provincial Key Laboratory of Intelligent Information Processing Shenzhen China

3D anomaly detection has recently become a significant focus in computer vision. Several advanced methods have achieved satisfying anomaly detection performance. However, they typically concentrate on the external structure of 3D samples and struggle to leverage the internal information embedded within samples. Inspired by the basic intuition of why not look inside for more, we introduce a straightforward method named Internal Spatial Modality Perception (ISMP) to explore the feature representation from internal views fully. Specifically, our proposed ISMP consists of a critical perception module, Spatial Insight Engine (SIE), which abstracts complex internal information of point clouds into essential global features. Besides, to better align structural information with point data, we propose an enhanced key point feature extraction module for amplifying spatial structure feature representation. Simultaneously, a novel feature filtering module is incorporated to reduce noise and redundant features for further aligning precise spatial structure. Extensive experiments validate the effectiveness of our proposed method, achieving object-level and pixel-level AUROC improvements of 3.2% and 13.1%, respectively, on the Real3D-AD benchmarks. Note that the strong generalization ability of SIE has been theoretically proven and is verified in both classification and segmentation tasks. Copyright © 2024, The Authors. All rights reserved.

关键词： Feature extraction

MASKED DISTILLATION WITH RECEPTIVE TOKENS

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Huang, Tao Zhang, Yuan You, Shan Wang, Fei Qian, Chen Cao, Jian Xu, Chang School of Computer Science Faculty of Engineering The University of Sydney Australia SenseTime Research Hong Kong School of Software and Microelectronics Peking University China University of Science and Technology of China China

Distilling from the feature maps can be fairly effective for dense prediction tasks since both the feature discriminability and localization priors can be well transferred. However, not every pixel contributes equally to the performance, and a good student should learn from what really matters to the teacher. In this paper, we introduce a learnable embedding dubbed receptive token to localize those pixels of interests (PoIs) in the feature map, with a distillation mask generated via pixel-wise attention. Then the distillation will be performed on the mask via pixel-wise reconstruction. In this way, a distillation mask actually indicates a pattern of pixel dependencies within feature maps of teacher. We thus adopt multiple receptive tokens to investigate more sophisticated and informative pixel dependencies to further enhance the distillation. To obtain a group of masks, the receptive tokens are learned via the regular task loss but with teacher fixed, and we also leverage a Dice loss to enrich the diversity of learned masks. Our method dubbed MasKD is simple and practical, and needs no priors of tasks in application. Experiments show that our MasKD can achieve state-of-the-art performance consistently on object detection and semantic segmentation benchmarks. Code is available at https://***/hunto/MasKD. © 2022, CC BY.

关键词： Distillation

Comprehensive Evaluation of ChatGPT Reliability Through Multilingual Inquiries

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Puttaparthi, Poorna Chander Reddy Deo, Soham Sanjay Gul, Hakan Tang, Yiming Shang, Weiyi Yu, Zhe Department of Software Engineering Rochester Institute of Technology RochesterNY United States School of Information Rochester Institute of Technology RochesterNY United States Department of Computer Science Rochester Institute of Technology RochesterNY United States Department of Electrical and Computer Engineering University of Waterloo WaterlooON Canada

ChatGPT is currently the most popular large language model (LLM), with over 100 million users, making a significant impact on people's lives. However, due to the presence of jailbreak vulnerabilities, ChatGPT might have negative effects on people's lives, potentially even facilitating criminal activities. Testing whether ChatGPT can cause jailbreak is crucial because it can enhance ChatGPT's security, reliability, and social responsibility. Inspired by previous research revealing the varied performance of LLMs in different language translations, we suspected that wrapping prompts in multiple languages might lead to ChatGPT jailbreak. To investigate this, we designed a study with a fuzzing testing approach to analyzing ChatGPT's cross-linguistic proficiency. Our study includes three strategies by automatically posing different formats of malicious questions to ChatGPT: (1) each malicious question involving only one language, (2) multilingual malicious questions, (3) specifying that ChatGPT responds in a language different from the questions. In addition, we also combine our strategies by utilizing prompt injection templates to wrap the three aforementioned types of questions. We examined a total of 7,892 Q&A data points, discovering that multilingual wrapping can indeed lead to ChatGPT's jailbreak, with different wrapping strategies having varying effects on jailbreak probability. Prompt injection can amplify the probability of jailbreak caused by multilingual wrapping. This study provides insights for OpenAI developers to enhance ChatGPT's support for language diversity and inclusion. © 2023, CC BY.

关键词：

Pipelined Training and Transmission for Federated Learning

学校读者我要写书评

暂无评论

Pipelined Training and Transmission for Federated Learning

International Conference on Parallel and Distributed Systems (ICPADS)

作者： Yihui Shen Hong Zhang Juncheng Jia Tao Deng Zhenhua Wang Zhiwei Teng Mianxiong Dong School of Computer Science and Technology Soochow University Suzhou China Collaborative Innovation Center of Novel Software Technology and Industrialization Nanjing China China Merchants New Intelligence Technology Co. Ltd. Beijing China Muroran Institute of Technology Muroran Japan

Federated learning is a distributed machine learning framework that enables distributed edge devices to jointly train machine learning models without exchanging private data. Because federated learning can leverage edge devices’ data and computing resources, the research on federated learning has set off an upsurge. However, one of major challenges faced by federated learning is the limited computation and communication resources of edge devices, which leads to low efficiency of federated learning. In this paper, we propose a federated learning framework FedPipeline, which focuses on the local training and uploading process to improve the convergence efficiency. Different from the existing federated learning algorithms with disjoint sequential steps of training and uploading for devices, FedPipeline exploits the parallelism of training and uploading that devices upload part of the model updates in advance while they are conducting local training simultaneously. In this way, the computation and communication resources of devices can be better utilized. We further design strategies for the scheduling of early uploading and the selection of uploaded updates to maximize the benefit of FedPipeline. Experimental results show that FedPipeline can effectively accelerate the convergence speed of federated learning and preserve the accuracy of the final model.

关键词：

General Virtual Images Construction Using Pixel Scrambling for Face Recognition 6th

学校读者我要写书评

暂无评论

General Virtual Images Construction Using Pixel Scrambling f...

6th International Conference on Artificial Intelligence and Security, ICAIS 2020

作者： Zhao, Yingnan Wu, Jie School of Computer and Software Nanjing University of Information Science and Technology Nanjing210044 China

ISBN: (纸本)9783030578831

Face recognition encounters the problem that multiple samples of the same object may be very different owing to the deformation of appearances. To synthesize reasonable virtual images of objects is a good way to solve it, which has been successfully used in face recognition. However, the schemes proposed are usually not suitable for other applications. In this paper, we propose a novel and general scheme to generate virtual images for deformable faces. We introduce the idea of pixel scrambling to our scheme and allow the pixels within an image block to be randomly rearranged to reflect possible variations of the appearance of objects. We demonstrate that virtual images obtained using pixel scrambling and original images are complementary in representing the appearance of objects. Extensive classification experiments on face databases show that the proposed virtual image scheme is very competent and can be combined with a number of classifiers to achieve surprising accuracy improvement. © 2020, Springer Nature Switzerland AG.

关键词： Face recognition

KnowAugNet: Multi-source medical knowledge augmented medication prediction network with multi-Level graph contrastive learning

学校读者我要写书评

暂无评论

arXiv 2022年

作者： An, Yang Jin, Bo Wei, Xiaopeng School of Software North University of China Taiyuan030051 China School of Innovation and Entrepreneurship Dalian University of Technology Dalian116024 China School of Computer Science and Technology Dalian University of Technology Dalian116024 China

Predicting medications is a crucial task in many intelligent healthcare systems. It can assist doctors in making informed medication decisions for patients according to electronic medical records (EMRs). However, medication prediction is a challenging data mining task due to the complex relations between medical codes. Most existing studies focus on utilizing inherent relations between homogeneous codes of medical ontology graph to enhance their representations using supervised methods, and few studies pay attention to the valuable relations between heterogeneous or homogeneous medical codes, e.g., synergistic relations between medications, concurrent relations between diseases, therapeutic relations between medication and disease from history EMRs, which further limits the prediction performance and application scenarios. Therefore, to address these limitations, this paper proposes KnowAugNet, a multi-sourced medical knowledge augmented medication prediction network which can fully capture the diverse relations between medical codes via multi-level graph contrastive learning framework. Specifically, KnowAugNet first leverages the graph contrastive learning using graph attention network as the encoder to capture the implicit relations between homogeneous medical codes from the medical ontology graph and obtains the knowledge augmented medical codes embedding vectors. Then, it utilizes the graph contrastive learning using a weighted graph convolutional network as the encoder to capture the correlative relations between homogeneous or heterogeneous medical codes from the constructed medical prior relation graph and obtains the relation augmented medical codes embedding vectors. Finally, the augmented medical codes embedding vectors and the supervised medical codes embedding vectors are retrieved and input to the sequential learning network to capture the temporal relations of medical codes and predict medications for patients. The experimental results demonstrate a co

关键词： Forecasting

From Misuse to Mastery: Enhancing Code Generation with Knowledge-Driven AI Chaining 23

学校读者我要写书评

暂无评论

From Misuse to Mastery: Enhancing Code Generation with Knowl...

Proceedings of the 38th IEEE/ACM International Conference on Automated software Engineering

作者： Xiaoxue Ren Xinyuan Ye Dehai Zhao Zhenchang Xing Xiaohu Yang School of Software Technology Zhejiang University Hangzhou China School of Computing Australian National University Canberra Australia CSIRO's Data61 Sydney Australia CSIRO's Data61 & Australian National University Sydney Australia College of Computer Science and Technology Zhejiang University Hangzhou China

ISBN: (纸本)9798350329964

Large Language Models (LLMs) have shown promising results in automatic code generation by improving coding efficiency to a certain extent. However, generating high-quality and reliable code remains a formidable task because of LLMs' lack of good programming practice, especially in exception handling. In this paper, we first conduct an empirical study and summarize three crucial challenges of LLMs in exception handling, i.e., incomplete exception handling, incorrect exception handling and abuse of try-catch. We then try prompts with different granularities to address such challenges, finding fine-grained knowledge-driven prompts works best. Based on our empirical study, we propose a novel Knowledge-driven Prompt Chaining-based code generation approach, name Kpc, which decomposes code generation into an AI chain with iterative check-rewrite steps and chains fine-grained knowledge-driven prompts to assist LLMs in considering exception-handling specifications. We evaluate our KPC-based approach with 3,079 code generation tasks extracted from the Java official API documentation. Extensive experimental results demonstrate that the KPC-based approach has considerable potential to ameliorate the quality of code generated by LLMs. It achieves this through proficiently managing exceptions and obtaining remarkable enhancements of 109.86% and 578.57% with static evaluation methods, as well as a reduction of 18 runtime bugs in the sampled dataset with dynamic validation.

关键词： large language model

Exploring Text-Guided Single Image Editing for Remote Sensing Images

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Han, Fangzhou Si, Lingyu Dong, Hongwei Zhang, Lamei Chen, Hao Du, Bo The Department of Information Engineering Harbin Institute of Technology Harbin China The National Key Laboratory of Space Integrated Information System Institute of Software Chinese Academy of Sciences Beijing China The Hubei Luojia Laboratory National Engineering Research Center for Multimedia Software School of Computer Science Wuhan University Wuhan China

Artificial intelligence generative content (AIGC) has significantly impacted image generation in the field of remote sensing. However, the equally important area of remote sensing image (RSI) editing has not received sufficient attention. Deep learning based editing methods generally involve two sequential stages: generation and editing. During the generation stage, consistency in content and details between the original and edited images must be maintained, while in the editing stage, controllability and accuracy of the edits should be ensured. For natural images, these challenges can be tackled by training generative backbones on large-scale benchmark datasets and using text guidance based on vision-language models (VLMs). However, these previously effective approaches become less viable for RSIs due to two reasons: First, existing generative RSI benchmark datasets do not fully capture the diversity of remote sensing scenarios, particularly in terms of variations in sensors, object types, and resolutions. Consequently, the generalization capacity of the trained backbone model is often inadequate for universal editing tasks on RSIs. Second, the large spatial resolution of RSIs exacerbates the problem in VLMs where a single text semantic corresponds to multiple image semantics, leading to the introduction of incorrect semantics when using text to guide RSI editing. To solve above problems, this paper proposes a text-guided RSI editing method that is controllable but stable, and can be trained using only a single image. It adopts a multi-scale training approach to preserve consistency without the need for training on extensive benchmark datasets, while leveraging RSI pre-trained VLMs and prompt ensembling (PE) to ensure accuracy and controllability in the text-guided editing process. High-quality RSI editing results can be achieved by first applying multi-scale training to the target image, followed by text-guided fine-tuning using RSI pre-trained VLMs and PE. Experi

关键词： Semantics