检索结果-内蒙古大学图书馆

11th International Conference on Learning Representations, ICLR 2023

作者： Wu, Hongqiu Liu, Yongxiang Shi, Hanwen Zhao, Hai Zhang, Min Department of Computer Science and Engineering Shanghai Jiao Tong University China Key Laboratory of Shanghai Education Commission for Intelligent Interaction and Cognitive Engineering Shanghai Jiao Tong University Shanghai China School of Computer Science and Technology Soochow University Suzhou China

Beyond the success story of adversarial training (AT) in the recent text domain on top of pre-trained language models (PLMs), our empirical study showcases the inconsistent gains from AT on some tasks, e.g. commonsense reasoning, named entity recognition. This paper investigates AT from the perspective of the contextualized language representation outputted by PLM encoders. We find the current AT attacks lean to generate sub-optimal adversarial examples that can fool the decoder part but have a minor effect on the encoder. However, we find it necessary to effectively deviate the latter one to allow AT to gain. Based on the observation, we propose simple yet effective Contextualized representation-Adversarial Training (CreAT), in which the attack is explicitly optimized to deviate the contextualized representation of the encoder. It allows a global optimization of adversarial examples that can fool the entire model. We also find CreAT gives rise to a better direction to optimize the adversarial examples, to let them less sensitive to hyperparameters. Compared to AT, CreAT produces consistent performance gains on a wider range of tasks and is proven to be more effective for language pretraining where only the encoder part is kept for downstream tasks. We achieve the new state-of-the-art performances on a series of challenging benchmarks, e.g. AdvGLUE (59.1 → 61.1), HellaSWAG (93.0 → 94.9), ANLI (68.1 → 69.3). © 2023 11th International Conference on Learning Representations, ICLR 2023. All rights reserved.

关键词： Global optimization

来源：评论

学校读者我要写书评

暂无评论

A 2D Geometry Based Grasping Pose Generation Algorithm for a Two-Finger Robot Hand

A 2D Geometry Based Grasping Pose Generation Algorithm for a...

引用

Iranian Conference on Electrical Engineering (ICEE)

作者： Arash Akbari Arman Akbari Mehdi Tale Masouleh Department of Computer Science University of Tehran Tehran Iran Human and Robot Interaction Laboratory School of Electrical and Computer Engineering Tehran Iran

Robot grasping is of paramount importance in industrial and service robotics. In recent years, various data-driven algorithms have been proposed to solve the problem of grasp detection and a part of them are based on reinforcement learning (RL) approaches. In a variety of proposed algorithms, random key points are being employed which will make the learning process inefficient and time-consuming. In this paper, a geometry-based algorithm is presented which can find grasp poses based on the geometry of the unknown object and propose the ones which may lead to successful grasping. For the grasp contacts computation part, the presented algorithm produces a finite number of key points based on the 2D shape of the object from a specific point of view. Afterward, it will narrow down the candidate points and output a finite number of successful grasp poses based on three grasp quality metrics for various unknown objects. Three approaches are proposed in order to achieve center points which can describe different parts of a 2D shape. Then, the obtained points are used as the center of circles which are tangent to the 2D shape contour. Also, a new grasp quality metric is proposed. The time of the grasp and the amount of object disorientation after grasping are considered as a metric to evaluate the successfulness of the grasp. Simulation results demonstrate that the proposed algorithm for unknown object grasping can find a finite number of successful grasp poses for different seen or unseen objects without using any random point,

关键词：

来源：评论

学校读者我要写书评

暂无评论

Transcriptomic Study of Neural Regulation in the Medial Prefrontal Cortex of Depressive Mice with Low-Intensity Focused Ultrasound 46

Transcriptomic Study of Neural Regulation in the Medial Pref...

引用

46th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBC 2024

作者： Cai, Qiuquan Meng, Lianghui Wang, Ling Ren, Jing Yang, Jiajia Ming, Dong Academy of Medical Engineering and Translational Medicine Tianjin University Tianjin 300072 China Brain-Computer Interaction and Human-Machine Fusion Haihe Laboratory Tianjin University Tianjin 300072 China Tianjin Key Laboratory of Brain Science and Neural Engineering Tianjin University Tianjin 300072 China

ISBN: (纸本)9798350371499

Exploring new effective treatments for depression holds important social significance and clinical value. Low-intensity focused ultrasound stimulation (LIFUS) has been proven to have significant neuroprotective effects on depression. However, the specific genes regulated by LIFUS in depressive mice remain unclear. We established a depression mice model using chronic restraint stress (CRS) and applied LIFUS to the medial prefrontal cortex (mPFC) following CRS. We explored the impact of modeling and treatment on gene expression changes through transcriptomics, revealing that the therapeutic effect of LIFUS may be associated with the improvement of the activity of calcium signaling pathway and synaptic plasticity. The study provides preclinical evidence and a theoretical basis for applying LIFUS for depression treatment. © 2024 IEEE.

关键词： Mammals

来源：评论

学校读者我要写书评

暂无评论

Daily Pain Prediction in Workplace Using Gaussian Processes

Daily Pain Prediction in Workplace Using Gaussian Processes

引用

16th International Joint Conference on Biomedical Engineering Systems and Technologies, BIOSTEC 2023

作者： Puri, Chetanya Keyaerts, Stijn Szymanski, Maxwell Godderis, Lode Verbert, Katrien Luca, Stijn Vanrumste, Bart KU Leuven Belgium Leuven Belgium Department ofPublic Health and Primary Care KU Leuven Leuven Belgium Department of Computer Science KU Leuven Belgium Human-computer Interaction research group Department of Computer Science KU Leuven Belgium Department of Data Analysis and Mathematical Modelling Ghent University Ghent Belgium

Work-related Musculoskeletal disorders (MSDs) account for 60% of sickness-related absences and even permanent inability to work in the Europe, Long term impacts of MSDs include "Pain chronification'' which is the transition of temporary pain into persistent pain. Preventive pain management can lower the risk of chronic pain. It is therefore important to appropriately assess pain in advance, which can assist a person in improving their fear of returning to work. In this study, we analysed pain data acquired over time by a smartphone application from a number of participants. We attempt to forecast a person's future pain levels based on his or her prior pain data. Due to the self-reported nature of the data, modelling daily pain is challenging due to the large number of missing values. For pain prediction modelling of a test subject, we employ a subset selection strategy that dynamically selects a closest subset of individuals from the training data. The similarity between the test subject and the training subjects is determined via dynamic time warping-based dissimilarity measure based on the time limited historical data until a given point in time. The pain trends of these selected subset subjects is more similar to that of the individual of interest. Then, we employ a Gaussian processes regression model for modelling the pain. We empirically test our model using a leave-one-subject-out cross validation to attain 20% improvement over state-of-the-art results in early prediction of pain. © 2023 by SCITEPRESS – science and Technology Publications, Lda.

关键词： Prediction models

来源：评论

学校读者我要写书评

暂无评论

JoTyMo: Joint Type and Movement Detection from RGB Images of Before and After Interacting with Articulated Objects

JoTyMo: Joint Type and Movement Detection from RGB Images of...

引用

RSI/ISM International Conference on Robotics and Mechatronics (ICROM)

作者： Ehsan Forootan Hamed Ghasemi Mehdi Tale Masouleh Ahmad Kalhor Behzad Moshiri Human and Robot Interaction Laboratory Electrical and Computer Engineering Department University of Tehran

This paper presents a CNN-based network architecture aimed to classify and detect joint types within Articulated objects, specifically the Push-P joints, P-joints, R-joints and objects lacking joints. The movement modeling of the detected objects is explored by processing pre and post interaction RGB images within the SAPIEN PartNet-Mobility dataset. In order to achieve this, an architecture is proposed to leverage consecutive CNN encoders based on the VGG architecture, in order to classify joints based on pre and post-interaction images. Additionally, in order to detect the effect-point and movement vector, a convolutional encoder is applied for each joint type. Moreover, extensive evaluation is made to showcase the results, achieving 96% accuracy in joint classification and 94% accuracy in regression on the considered dataset. Elaborate details regarding network architectures, training procedures, and testing methodologies are available at source code repository.

关键词：

来源：评论

学校读者我要写书评

暂无评论

PROM: A Phrase-level Copying Mechanism with Pre-training for Abstractive Summarization 30

PROM: A Phrase-level Copying Mechanism with Pre-training for...

引用

Joint 30th International Conference on Computational Linguistics and 14th International Conference on Language Resources and Evaluation, LREC-COLING 2024

作者： Ma, Xinbei Gong, Yeyun He, Pengcheng Zhao, Hai Duan, Nan Department of Computer Science and Engineering Shanghai Jiao Tong University China Key Laboratory of Shanghai Education Commission for Intelligent Interaction and Cognitive Engineering Shanghai Jiao Tong University China Microsoft Research Asia China

ISBN: (纸本)9782493814104

Based on the remarkable achievements of pre-trained language models in abstractive summarization, the copying mechanism has proved helpful by improving the factuality, stability, and overall performance. This work proposes PROM, a new PhRase-level cOpying Mechanism that enhances attention on n-grams, which can be applied to zero-shot summarization with pre-training. PROM adds an indicator layer to explicitly pick up tokens in n-gram that can be copied from the source, and calculates an auxiliary loss for the copying prediction. Empirical studies show that PROM makes significant improvements in fine-tuning on benchmarks. In the zero-shot setting, PROM is utilized in the self-supervised pre-training on raw corpora and provides new general baselines on a wide range of summarization datasets. Further analysis shows that PROM performs more reasonable copying and contributes to faithfulness. Our code is publicly available at https://***/xbmxb/PROM. © 2024 ELRA Language Resource Association: CC BY-NC 4.0.

关键词： Computational linguistics

来源：评论

学校读者我要写书评

暂无评论

Enhanced Fine-Tuning of Lightweight Domain-Specific Q&A Model Based on Large Language Models 35

Enhanced Fine-Tuning of Lightweight Domain-Specific Q&A Mode...

引用

35th IEEE International Symposium on Software Reliability Engineering Workshops, ISSREW 2024

作者： Zhang, Shenglin Zhu, Pengtian Ma, Minghua Wang, Jiagang Sun, Yongqian Li, Dongwen Wang, Jingyu Guo, Qianying Hua, Xiaolei Zhu, Lin Pei, Dan Nankai University China Haihe Laboratory of Information Technology Application Innovation China Microsoft Tsinghua University China Tianjin Key Laboratory of Software Experience and Human Computer Interaction China China Mobile Research Institute China Beijing National Research Center for Information Science and Technology China

ISBN: (纸本)9798350367041

Large language models (LLMs) excel at general question-answering (Q&A) but often fall short in specialized domains due to a lack of domain-specific knowledge. Commercial companies face the dual challenges of privacy protection and resource constraints when involving LLMs for fine-tuning. This paper propose a novel framework, Self-Evolution, designed to address these issues by leveraging lightweight open-source LLMs through multiple iterative fine-tuning rounds. To enhance the efficiency of iterative fine-tuning, Self-Evolution employ a strategy that filters and reinforces the knowledge with higher value during the iterative process. We employed Self-Evolution on Qwen1.5-7B-Chat using 4,000 documents containing rich domain knowledge from China Mobile, achieving a performance score 174% higher on domain-specific question-answering evaluations than Qwen1.5-7B-Chat and even 22% higher than Qwen1.5-72B-Chat. Self-Evolution has been deployed in China Mobile's daily operation and maintenance for 117 days, and it improves the efficiency of locating alarms, fixing problems, and finding related reports, with an average efficiency improvement of over 18.6%. In addition, we release Self-Evolution framework code in https://***/Zero-Pointer/Self-Evolution. © 2024 IEEE.

关键词： Data mining

来源：评论

学校读者我要写书评

暂无评论

Tiles to Move: Investigating Tile-Based Locomotion for Virtual Reality

引用

Proceedings of the ACM on human-computer interaction 2023年第CHI PLAY期7卷 1044–1067页

作者： Funke, Jana Franceska Schikorr, Anja Karaosmanoglu, Sukran Hirzle, Teresa Steinicke, Frank Rukzio, Enrico Institute of Media Informatics Ulm University James-Franck Ring Ulm Ulm89081 Germany Human-Computer Interaction Universität Hamburg Hamburg Vogt-Kölln-Straße 30 Hamburg22527 Germany Department of Computer Science University of Copenhagen Copenhagen Universitetsparken 1 Copenhagen2100 Denmark Institute of Media Informatics Ulm University Ulm Germany

Tile-based locomotion (TBL) is a popular locomotion technique for computer, console, and board games. However, despite its simplicity and unconventional movement, the transfer of TBL to virtual reality (VR) as a game platform remains unexplored. To fill this gap, we introduce TBL for VR on the example of two techniques: a controller and a feet-based one. In a first user study, we evaluated the usability and acceptance of the techniques compared to teleportation and touchpad locomotion. In a second exploratory user study, we evaluated the user experience of both TBL techniques in a maze and a museum scenario. The findings show that both techniques provide enjoyment and acceptable usability by creating either a relaxing (controller-based) or a physically active (feet-based) solution. Finally, our results highlight that TBL techniques work particularly well for small, constrained spaces that allow users to focus on exploring details in the nearby environment (important for games) in contrast to large open spaces that require faster locomotion, like teleportation. © 2023 Owner/Author.

关键词： Virtual reality

来源：评论

学校读者我要写书评

暂无评论

Decentralized Federated Graph Learning with Lightweight Zero Trust Architecture for Next-Generation Networking Security

引用

IEEE Journal on Selected Areas in Communications 2025年第6期43卷 1908-1922页

作者： Zhou, Xiaokang Liang, Wei Wang, Kevin I-Kai Yada, Katsutoshi Yang, Laurence T. Ma, Jianhua Jin, Qun Kansai University Faculty of Business Data Science Osaka565-0823 Japan RIKEN Center for Advanced Intelligence Project RIKEN Tokyo103-0027 Japan Hunan University of Technology and Business School of Computer Science Changsha410205 China Xiangjiang Laboratory Changsha410205 China The University of Auckland Department of Electrical Computer and Software Engineering Auckland1010 New Zealand St. Francis Xavier University Department of Computer Science AntigonishNSB2G 2W5 Canada Hosei University Faculty of Computer and Information Sciences Chiyoda-ku102-8160 Japan Waseda University Faculty of Human Sciences Tokorozawa359-1192 Japan

The rapid development and usage of digital technologies in modern intelligent systems and applications bring critical challenges on data security and privacy. It is essential to allow cross-organizational data sharing to achieve smart service provisioning, while preventing unauthorized access and data leak to ensure end users' efficient and secure collaborations. Federated Learning (FL) offers a promising pathway to enable innovative collaboration across multiple organizations. However, more stringent security policies are needed to ensure authenticity of participating entities, safeguard data during communication, and prevent malicious activities. In this paper, we propose a Decentralized Federated Graph Learning (FGL) with Lightweight Zero Trust Architecture (ZTA) model, named DFGL-LZTA, to provide context-aware security with dynamic defense policy update, while maintaining computational and communication efficiency in resource-constrained environments, for highly distributed and heterogeneous systems in next-generation networking. Specifically, with a re-designed lightweight ZTA, which leverages adaptive privacy preservation and reputation-based aggregation together to tackle multi-level security threats (e.g., data-level, model-level, and identity-level attacks), a Proximal Policy Optimization (PPO) based Deep Reinforcement Learning (DRL) agent is introduced to enable the real-time and adaptive security policy update and optimization based on contextual features. A hierarchical Graph Attention Network (GAT) mechanism is then improved and applied to facilitate the dynamic subgraph learning in local training with a layer-wise architecture, while a so-called sparse global aggregation scheme is developed to balance the communication efficiency and model robustness in a P2P manner. Experiments and evaluations conducted based on two open-source datasets and one synthetic dataset demonstrate the usefulness of our proposed model in terms of training performance, computa

关键词： Federated learning

来源：评论

学校读者我要写书评

暂无评论

Experimental Study on Automatically Assembling Custom Catering Packages With a 3-DOF Delta Robot Using Deep Learning Methods

Experimental Study on Automatically Assembling Custom Cateri...

引用

Iranian Conference on Electrical Engineering (ICEE)

作者： Reihaneh Yourdkhani Arash Tavoosian Navid Asadi Khomami Mehdi Tale Masouleh Electrical and Computer Engineering Department Human and Robot Interaction Laboratory University of Tehran Mechanical Engineering Department University of Tehran

ISBN: (数字)9798350376340

ISBN: (纸本)9798350376357

This paper introduces a pioneering experimental study on the automated packing of a catering package using a two-fingered gripper affixed to a 3-degree-of-freedom Delta parallel robot. A distinctive contribution lies in the application of a deep learning approach to tackle this challenge. A custom dataset, comprising 1,500 images, is meticulously curated for this endeavor, representing a noteworthy initiative as the first dataset focusing on Persian-manufactured products. The study employs the YOLOV5 model for object detection, followed by segmentation using the FastSAM model. Subsequently, rotation angle calculation is facilitated with segmentation masks, and a rotated rectangle encapsulating the object is generated. This rectangle forms the basis for calculating two grasp points using a novel geometrical approach involving eigenvectors. An extensive experimental study validates the proposed model, where all pertinent information is seamlessly transmitted to the 3-DOF Delta parallel robot. The proposed algorithm ensures real-time detection, calibration, and the fully autonomous packing process of a catering package, boasting an impressive over 80% success rate in automatic grasping. This study marks a significant stride in advancing the capabilities of robotic systems for practical applications in packaging automation.

关键词： Deep learning YOLO Parallel robots Image segmentation 3-DOF Focusing Grasping

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：