检索结果-内蒙古大学图书馆

arXiv 2022年

作者： Wan, Zhongwei Yin, Yichun Zhang, Wei Shi, Jiaxin Shang, Lifeng Chen, Guangyong Jiang, Xin Liu, Qun Guangdong Provincial Key Laboratory of Computer Vision and Virtual Reality Technology Shenzhen Institute of Advanced Technology Chinese Academy of Science China University of Chinese Academy of Sciences China Huawei Noah's Ark Lab Hong Kong Huawei Cloud Computing Zhejiang Lab China

Recently, domain-specific PLMs have been proposed to boost the task performance of specific domains (e.g., biomedical and computer science) by continuing to pre-train general PLMs with domain-specific corpora. However, this Domain-Adaptive Pre-Training (DAPT;Gururangan et al. (2020)) tends to forget the previous general knowledge acquired by general PLMs, which leads to a catastrophic forgetting phenomenon and sub-optimal performance. To alleviate this problem, we propose a new framework of General Memory- Augmented Pre-trained Language Model (GMAP), which augments the domain-specific PLM by a memory representation built from the frozen general PLM without losing any general knowledge. Specifically, we propose a new memory-augmented layer, and based on it, different augmented strategies are explored to build the memory representation and then adaptively fuse it into the domain-specific PLM. We demonstrate the effectiveness of G-MAP on various domains (biomedical and computer science publications, news, and reviews) and different kinds (text classification, QA, NER) of tasks, and the extensive results show that the proposed G-MAP1 can achieve SOTA results on all tasks. © 2022, CC BY.

关键词： Computational linguistics

来源：评论

学校读者我要写书评

暂无评论

Deep semi-supervised knowledge distillation for overlapping cervical cell instance segmentation 23rd

Deep semi-supervised knowledge distillation for overlapping ...

引用

23rd International Conference on Medical Image Computing and computer-Assisted Intervention, MICCAI 2020

作者： Zhou, Yanning Chen, Hao Lin, Huangjing Heng, Pheng-Ann Department of Computer Science and Engineering The Chinese University of Hong Kong Hong Kong Guangdong Provincial Key Laboratory of Computer Vision and Virtual Reality Technology Shenzhen Institutes of Advanced Technology Chinese Academy of Sciences Shenzhen China

ISBN: (纸本)9783030597092

Deep learning methods show promising results for overlapping cervical cell instance segmentation. However, in order to train a model with good generalization ability, voluminous pixel-level annotations are demanded which is quite expensive and time-consuming for acquisition. In this paper, we propose to leverage both labeled and unlabeled data for instance segmentation with improved accuracy by knowledge distillation. We propose a novel Mask-guided Mean Teacher framework with Perturbation-sensitive Sample Mining (MMT-PSM), which consists of a teacher and a student network during training. Two networks are encouraged to be consistent both in feature and semantic level under small perturbations. The teachers self-ensemble predictions from K-time augmented samples are used to construct the reliable pseudo-labels for optimizing the student. We design a novel strategy to estimate the sensitivity to perturbations for each proposal and select informative samples from massive cases to facilitate fast and effective semantic distillation. In addition, to eliminate the unavoidable noise from the background region, we propose to use the predicted segmentation mask as guidance to enforce the feature distillation in the foreground region. Experiments show that the proposed method improves the performance significantly compared with the supervised method learned from labeled data only, and outperforms state-of-the-art semi-supervised methods. Code: https://***/SIAAAAAA/MMT-PSM. © Springer Nature Switzerland AG 2020.

关键词： Distillation

来源：评论

学校读者我要写书评

暂无评论

Model Ownership Protection for Healthcare Consumer Electronics in Federated Edge Learning

引用

IEEE Transactions on Consumer Electronics 2024年

作者： Wang, Rong Liang, Junchuan Feng, Chaosheng Chakraborty, Chinmay Alqahtani, Fayez Yu, Keping Sichuan Normal University College of Computer Science Chengdu610101 China Sichuan Normal University Visual Computing and Virtual Reality Key Laboratory of Sichuan Chengdu610068 China Birla Institute of Technology Mesra India King Saud University College of Computer and Information Sciences Software Engineering Department Riyadh12372 Saudi Arabia Hosei University Graduate School of Science and Engineering Tokyo184-8584 Japan

With the rising demand for intelligent services and privacy protection in consumer artificial intelligence (AI), federated edge learning has emerged as a beacon for privacy-preserving distributed machine learning. This non-centralized approach offers distinct data privacy advantages, making it especially suitable for applications with strict privacy policies, such as consumer-centric healthcare systems. However, consumer electronics/devices in federated edge learning might unlawfully violate model ownership by duplicating or distributing models. Another concern is that a central server, if compromised, can become a conduit for model leakage, whether for profit motives or due to external intrusions, thus exacerbating the model ownership violation. To our knowledge, few works have focused on model ownership protection for federated edge learning in the consumer-centric healthcare domain. In this paper, we introduce a federated model ownership protection (FedMOP) approach. FedMOP is a model ownership safeguard and a formidable fortress built upon two core innovations. First, FedMOP can swiftly embed and replace watermarks during local training, allowing for the accurate identification and subsequent flagging of deceitful consumer devices. Second, FedMOP ensures that the server only accesses encrypted versions of the federated model, keeping the model a secret. Extensive evaluations demonstrate the robustness and effectiveness of FedMOP. We hope that FedMOP can light the way for surmounting similar ethical challenges in the broader realm of consumer-edge AI. © 1975-2011 IEEE.

关键词： ethical AI Federated edge learning healthcare consumer electronicss model ownership protection

来源：评论

学校读者我要写书评

暂无评论

ReAct: Temporal Action Detection with Relational Queries

arXiv

引用

arXiv 2022年

作者： Shi, Dingfeng Zhong, Yujie Cao, Qiong Zhang, Jing Ma, Lin Li, Jia Tao, Dacheng State Key Laboratory of Virtual Reality Technology and Systems School of Computer Science and Engineering Beihang University China Meituan Inc China JD Explore Academy China The University of Sydney Australia

This work aims at advancing temporal action detection (TAD) using an encoder-decoder framework with action queries, similar to DETR, which has shown great success in object detection. However, the framework suffers from several problems if directly applied to TAD: the insufficient exploration of inter-query relation in the decoder, the inadequate classification training due to a limited number of training samples, and the unreliable classification scores at inference. To this end, we first propose a relational attention mechanism in the decoder, which guides the attention among queries based on their relations. Moreover, we propose two losses to facilitate and stabilize the training of action classification. Lastly, we propose to predict the localization quality of each action query at inference in order to distinguish high-quality queries. The proposed method, named ReAct, achieves the state-of-the-art performance on THUMOS14, with much lower computational costs than previous methods. Besides, extensive ablation studies are conducted to verify the effectiveness of each proposed component. The code is available at https://***/sssste/React. © 2022, CC BY.

关键词： Object detection

来源：评论

学校读者我要写书评

暂无评论

Bold but Cautious: Unlocking the Potential of Personalized Federated Learning through Cautiously Aggressive Collaboration

Bold but Cautious: Unlocking the Potential of Personalized F...

引用

International Conference on computer vision (ICCV)

作者： Xinghao Wu Xuefeng Liu Jianwei Niu Guogang Zhu Shaojie Tang State Key Laboratory of Virtual Reality Technology and Systems School of Computer Science and Engineering Beihang University Beijing China Zhongguancun Laboratory Beijing China Zhengzhou University Research Institute of Industrial Technology School of Information Engineering Zhengzhou University Zhengzhou China Jindal School of Management University of Texas at Dallas USA

Personalized federated learning (PFL) reduces the impact of non-independent and identically distributed (non-IID) data among clients by allowing each client to train a personalized model when collaborating with others. A key question in PFL is to decide which parameters of a client should be localized or shared with others. In current mainstream approaches, all layers that are sensitive to non-IID data (such as classifier layers) are generally personalized. The reasoning behind this approach is understandable, as localizing parameters that are easily influenced by non-IID data can prevent the potential negative effect of collaboration. However, we believe that this approach is too conservative for collaboration. For example, for a certain client, even if its parameters are easily influenced by non-IID data, it can still benefit by sharing these parameters with clients having similar data distribution. This observation emphasizes the importance of considering not only the sensitivity to non-IID data but also the similarity of data distribution when determining which parameters should be localized in PFL. This paper introduces a novel guideline for client collaboration in PFL. Unlike existing approaches that prohibit all collaboration of sensitive parameters, our guideline allows clients to share more parameters with others, leading to improved model performance. Additionally, we propose a new PFL method named FedCAC, which employs a quantitative metric to evaluate each parameter’s sensitivity to non-IID data and carefully selects collaborators based on this evaluation. Experimental results demonstrate that FedCAC enables clients to share more parameters with others, resulting in superior performance compared to state-of-the-art methods, particularly in scenarios where clients have diverse distributions. The code is integrated into our FL training framework: https://***/kxzxvbk/Fling.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Enabling Communication-Efficient Federated Learning via Distributed Compressed Sensing

Enabling Communication-Efficient Federated Learning via Dist...

引用

IEEE Annual Joint Conference: INFOCOM, IEEE computer and Communications Societies

作者： Yixuan Guan Xuefeng Liu Tao Ren Jianwei Niu State Key Laboratory of Virtual Reality Technology and Systems School of Computer Science and Engineering Beihang University Beijing China Zhongguancun Laboratory Beijing China Laboratory for Internet Software Technologies Institute of Software Chinese Academy of Sciences Beijing China Zhengzhou University Research Institute of Industrial Technology School of Information Engineering Zhengzhou University Zhengzhou China

Federated learning (FL) trains a shared global model by periodically aggregating gradients from local devices. Communication overhead becomes a principal bottleneck in FL since participating devices usually suffer from limited bandwidth and unreliable connections in uplink transmission. To address this problem, the gradient compression methods based on compressed sensing (CS) theory have been put forward recently. However, most existing CS-based works compress gradients independently, ignoring the gradient correlations between participants or adjacent communication rounds, which constrains the achievement of higher compression rates. In view of the above observation, we propose a novel gradient compression scheme named FedDCS, guided by distributed compressed sensing (DCS) theory. Following the design philosophy of separate encoding and joint decoding in DCS, FedDCS compresses gradients for participants in each round separately while reconstructing them at the central server jointly via fully exploiting correlated gradients from the previous round, which are known as side information (SI). Benefiting from this design, reconstruction performance is significantly improved with fewer decoding errors also iterations under the identical compression rate, and the total uploading bits to achieve model convergence are considerably reduced. Theoretical analysis and extensive experiments conducted on MNIST and Fashion-MNIST both verify the effectiveness of our approach.

关键词：

来源：评论

学校读者我要写书评

暂无评论

FEAT: Towards Fast Environment-Adaptive Task Offloading and Power Allocation in MEC

FEAT: Towards Fast Environment-Adaptive Task Offloading and ...

引用

IEEE Annual Joint Conference: INFOCOM, IEEE computer and Communications Societies

作者： Tao Ren Zheyuan Hu Hang He Jianwei Niu Xuefeng Liu Laboratory for Internet Software Technologies Institute of Software Chinese Academy of Sciences Beijing China State Key Laboratory of Virtual Reality Technology and Systems School of Computer Science and Engineering Beihang University Beijing China Zhongguancun Laboratory Beijing China Zhengzhou University Research Institute of Industrial Technology School of Information Engineering Zhengzhou University Zhengzhou China

Mobile edge computing (MEC) has been proposed to provide mobile devices with both satisfactory computing resources and latency. key issues in MEC include task offloading and power allocation (TOPA), for which deep reinforcement learning (DRL) is becoming a popular methodology. However, most DRL-based TOPA approaches are typically developed in a certain environment, lacking the adaptability to unseen environments. Motivated by this, this paper proposes a Fast Environment-Adaptive TOPA (FEAT) approach that could adapt to unseen environments with little fine-tuning. Specifically, we first split MEC states into the internal state and environmental state. Then, based on these two types of states, we develop two main components of FEAT — a group of internal state-dependent TOPA meta-policies and an environmental state-embedded steerer. Meta-policies learn TOPA skills within the internal state space (allowing reusing meta-policies in different environments), while the steerer learns to choose appropriate meta-policies according to embedded environmental states. When encountering an unseen environment with the same internal state space, FEAT only needs to fine-tune the steerer using the newly embedded environmental state with few internal state explorations. Extensive experimental results on simulation and testbed demonstrate that FEAT outperforms the state-of-the-art by more than 16.4% in terms of fine-tuning speeds.

关键词：

来源：评论

学校读者我要写书评

暂无评论

TransOff: Towards Fast Transferable Computation Offloading in MEC via Embedded Reinforcement Learning

TransOff: Towards Fast Transferable Computation Offloading i...

引用

International Conference on Distributed Computing Systems

作者： Zheyuan Hu Jianwei Niu Tao Ren State Key Laboratory of Virtual Reality Technology and Systems School of Computer Science and Engineering Beihang University Beijing China Zhengzhou University Research Institute of Industrial Technology School of Information Engineering Zhengzhou University Zhengzhou China Laboratory for Internet Software Technologies Institute of Software Chinese Academy of Sciences Beijing China

Mobile edge computing (MEC) has been proposed as a promising paradigm to provide mobile devices with both satisfactory computing capacity and task latency. One key issue in MEC is computation offloading (CompOff), which has attracted numerous research interests. Most existing CompOff approaches are developed based on iterative programming (IterProg), that calculates a CompOff action based on system dynamics each time mobile tasks arrive. Due to the heavy dependency of IterProg on reliable system dynamics, as well as the online computational burden, recent years have seen a popular trend to develop CompOff approaches based on deep reinforcement learning (DRL), which could generate real-time model-free CompOff actions. However, due to the intrinsic poor generalization of DRL, it is hard to directly apply DRL-based policies in new MEC environments, and long-time fine-tuning is often required. To address the challenge, this paper proposes a fast transferable CompOff framework (named TransOff), based on the idea of embedded reinforcement learning. Specifically, TransOff is composed of multiple primitive CompOff policies (pCOPs) and a multiplicative composition function (MCF). The pCOPs and MCF are pre-trained in a diverse variety of MEC environments. When encountering new MEC environments, pCOPs are kept fixed to prevent catastrophic forgetting of pre-trained CompOff skills, while only MCF is fine-tuned to produce new compositions of pCOPs to achieve fast transfer. We conduct extensive experiments via both numerical simulation and real testbed, indicating the fast transfer ability of TransOff compared to the state-of-the-art DRL-based and meta learning-based CompOff approaches.

关键词：

来源：评论

学校读者我要写书评

暂无评论

RePFormer: Refinement Pyramid Transformer for Robust Facial Landmark Detection

arXiv

引用

arXiv 2022年

作者： Li, Jinpeng Jin, Haibo Liao, Shengcai Shao, Ling Heng, Pheng-Ann Department of Computer Science and Engineering The Chinese University of Hong Kong Hong Kong Department of Computer Science and Engineering The Hong Kong University of Science and Technology Hong Kong United Arab Emirates Terminus Group China Guangdong Provincial Key Laboratory of Computer Vision and Virtual Reality Technology Shenzhen Institutes of Advanced Technology Chinese Academy of Sciences Shenzhen China

This paper presents a Refinement Pyramid Transformer (RePFormer) for robust facial landmark detection. Most facial landmark detectors focus on learning representative image features. However, these CNN-based feature representations are not robust enough to handle complex real-world scenarios due to ignoring the internal structure of landmarks, as well as the relations between landmarks and context. In this work, we formulate the facial landmark detection task as refining landmark queries along pyramid memories. Specifically, a pyramid transformer head (PTH) is introduced to build both homologous relations among landmarks and heterologous relations between landmarks and cross-scale contexts. Besides, a dynamic landmark refinement (DLR) module is designed to decompose the landmark regression into an end-to-end refinement procedure, where the dynamically aggregated queries are transformed to residual coordinates predictions. Extensive experimental results on four facial landmark detection benchmarks and their various subsets demonstrate the superior performance and high robustness of our framework. Copyright © 2022, The Authors. All rights reserved.

关键词： Benchmarking

来源：评论

学校读者我要写书评

暂无评论

VS-Net: A Voxel Encoding and Sparse Convolution Embedded Network for LiDAR 3D Object Detection

VS-Net: A Voxel Encoding and Sparse Convolution Embedded Net...

引用

2022 IEEE International Conference on Robotics and Biomimetics, ROBIO 2022

作者： Liu, Meng Niu, Jianwei Liu, Yu The China North Artificial Intelligence and Innovation Research Institute Beijing100072 China The China North Vehicle Research Institute Beijing100072 China The State Key Laboratory of Software Development Environment School of Computer Science and Engineering Beihang University Beijing100191 China The State Key Laboratory of Virtual Reality Technology and Systems School of Computer Science and Engineering Beihang University Beijing100191 China The Hangzhou Innovation Institute Beihang University Hangzhou310051 China The Zhengzhou University Research Institute of Industrial Technology School of Information Engineering Zhengzhou University Zhengzhou450001 China

ISBN: (纸本)9781665481090

Accurate and efficient 3D object detection in LiDAR point cloud is critical for autonomous driving. In this paper, we provide a solution for LiDAR-only detection which employs voxel features and a sparse convolution network. Specifically, we utilized voxel centers to encode point clouds' voxel features. Then, combining the sparse convolution method, we leveraged residual learning to design the backbone network. The detector output predicted results for multi-category and multi-object detection. Our method was evaluated on three popular and challenging datasets (KITTI, nuScence and Waymo). Experimental results demonstrated that the accuracy and speed (27.8 FPS) of our model could effectively support object detection for autonomous driving in various scenarios. © 2022 IEEE.

关键词： Object detection

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：