检索结果-内蒙古大学图书馆

1st International Conference on Smart Technologies Communication and Robotics, STCR 2021

作者： Manochitra, S. Partheepan, Ravinthiran Kalasalingam Institute of Technology Faculty of Computer Science and Engineering Tamilnadu Virudhunagar India Vytautas Magnus University Faculty of Informatics Master Programme - Applied Informatics Kaunas Lithuania

ISBN: (纸本)9781665418065

Emotions plays a potential role in human computer interaction which are having an obligatory models of cognitive measures. The emotions are dominated by the human physiological communication channels. Those emotions could affect several human activities like learning, decision making, communication, perception, cognitive and so on. Also, some of the emotions are quite difficult to measure and interpret and having difficulty in incorporating them into a technology and design. On the other hand, state of the technologies has enabled us to capture the emotions and integrate them for the use of human computer interaction contexts. This paper delivers an attempt to model the emotions using electroencephalography (EEG). The video stimuli for the two representatives are based on a few negative, positive and neutral contents were shown to volunteers and EEG data was obtained. Further, EEG analysis were experimented and collected about their emotional states with subject's self reports. This study determines that the human emotions could be modelled for affect assessment module or for the usage in human computer interaction and affect based intelligent interactions. © 2021 IEEE.

关键词： Human computer interaction

来源：评论

学校读者我要写书评

暂无评论

Purify unlearnable examples via rate-constrained variational autoencoders 24

Purify unlearnable examples via rate-constrained variational...

引用

Proceedings of the 41st International Conference on Machine Learning

作者： Yi Yu Yufei Wang Song Xia Wenhan Yang Shijian Lu Yap-Peng Tan Alex C. Kot Rapid-Rich Object Search Lab Interdisciplinary Graduate Programme and School of Electrical and Electronic Engineering Nanyang Technological University Singapore School of Electrical and Electronic Engineering Nanyang Technological University Singapore PengCheng Laboratory Shenzhen China School of Computer Science and Engineering Nanyang Technological University Singapore

Unlearnable examples (UEs) seek to maximize testing error by making subtle modifications to training examples that are correctly labeled. Defenses against these poisoning attacks can be categorized based on whether specific interventions are adopted during training. The first approach is training-time defense, such as adversarial training, which can mitigate poisoning effects but is computationally intensive. The other approach is pre-training purification, e.g., image short squeezing, which consists of several simple compressions but often encounters challenges in dealing with various UEs. Our work provides a novel disentanglement mechanism to build an efficient pre-training purification method. Firstly, we uncover rate-constrained variational autoencoders (VAEs), demonstrating a clear tendency to suppress the perturbations in UEs. We subsequently conduct a theoretical analysis for this phenomenon. Building upon these insights, we introduce a disentangle variational autoencoder (DVAE), capable of disentangling the perturbations with learnable class-wise embeddings. Based on this network, a two-stage purification approach is naturally developed. The first stage focuses on roughly eliminating perturbations, while the second stage produces refined, poison-free results, ensuring effectiveness and robustness across various scenarios. Extensive experiments demonstrate the remarkable performance of our method across CIFAR-10, CIFAR-100, and a 100-class ImageNet-subset. Code is available at https://***/yuyi-sd/D-VAE.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Developing Critical Thinking Practices Interwoven with Generative AI Usage in an Introductory Programming Course

Developing Critical Thinking Practices Interwoven with Gener...

引用

IEEE Education engineering (EDUCON)

作者： Arne Styve Outi T. Virkki Usman Naeem Department of ICT and Natural Sciences Norwegian University of Science and Technology (NTNU) Ålesund Norway Information Technology Degree Programme Haaga-Helia University of Applied Sciences Helsinki Finland School of Electronic Engineering and Computer Science Queen Mary University of London London UK

ISBN: (数字)9798350394023

ISBN: (纸本)9798350394030

Software development has evolved significantly. In the past, developers were required to have comprehensive understanding of programming languages, algorithms, and computer architecture. However, with the emergence of the Internet, software libraries, frameworks, and forums became widely available, which utilize reusable software components that can reduce development time and costs. The advent of Generative Artificial Intelligence (AI) tools, such as ChatGPT, GitHub Copilot, and Amazon CodeWhisperer, has further enhanced the developer's toolkit, as these tools can be used for a wide variety of tasks such as code generation, documentation, commenting and reviewing. As programming is often slow and requires trial and error, novice programmers can be tempted to apply the first solution found on the Internet or proposed by an AI tool without much critical reflection or notion of responsibility. Hence, the advances of AI have raised both excitement and concerns among Information Technology (IT)/computer science (CS) students and educators. Yet, AI tools are here to stay, and students must learn to use them responsibly. The aim of this paper is to investigate how to design learning activities that introduce Generative AI tools (GitHub Copilot and ChatGPT) for programming while promoting critical thinking practices among students in an introductory programming course in the first semester. Students' opinions and customs were surveyed before and after the AI-based programming assignment. The results indicate that students' awareness of the possibilities and limitations of AI, as well as practices of critical thinking in programming increased. This is encouraging as critical thinking is an integral part of best programming practices.

关键词： Software libraries Generative AI Software algorithms Chatbots Software Reflection Internet

来源：评论

学校读者我要写书评

暂无评论

Computation-Efficient Model Predictive Control of a Multilevel Current Source Converter

Computation-Efficient Model Predictive Control of a Multilev...

引用

Annual Conference of Industrial Electronics Society

作者： Muyu Chen Zhige Yuan Amer M. Y. M. Ghias Vun Chan Hua Nicholas School of Electrical and Electronic Engineering Collab. Initiative Interdisciplinary Graduate Programme Nanyang Technological University Singapore School of Electrical and Electronic Engineering Nanyang Technological University Singapore School of Computer Science and Engineering Nanyang Technological University Singapore

This paper presents a modified finite control set model predictive control (FCS-MPC) that can significantly reduce computation time and maintain good quality of output voltage. The switching states of a three-phase five-level current source converter (5L-CSC) are sorted according to the corresponding output converter current level. The influences of every switching state on the converter inductor current are analyzed. Then the optimal switching state can be anchored based on the converter current level that minimizes the cost function and converter inductor current conditions. The proposed method eliminates the need for the discrete-time model establishment for the converter. It avoids the ergodic search through all switching states, easing the controller's computation burden. Besides, this algorithm achieves multiple control goals without assigning different weighting factors to various control objectives. The simulated output voltage total harmonics distortion (THD), the inductor current ripple, and the execution time of the modified model predictive control and traditional model predictive control are compared to validate the effectiveness of the proposed algorithm.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Room-Temperature Switching of Perpendicular Magnetization by Magnon Torques

引用

Physical Review Applied 2023年第3期19卷 034039-034039页

作者： Guoyi Shi Fei Wang Hui Ru Tan Shishun Zhao Yakun Liu Dongsheng Yang Kyusup Lee Yuchen Pu Shuhan Yang Anjan Soumyanarayanan Hyunsoo Yang Integrative Sciences and Engineering Programme NUS Graduate School National University of Singapore Singapore 119077 Singapore Department of Electrical and Computer Engineering National University of Singapore Singapore 117576 Singapore Institute of Materials Research & Engineering Agency for Science Technology & Research (A*STAR) Singapore 138634 Singapore Department of Physics National University of Singapore Singapore 117551 Singapore

Electron-mediated spin torque provides a fast and efficient method to manipulate magnetization; however, electron motion inevitably brings about the generation of Joule heat and corresponding power consumption. Magnon-mediated spin torque, without involving moving electrons, could circumvent the energy dissipation issue. In this work, we fabricate a sandwich structure of topological insulator/antiferromagnetic insulator/ferromagnet with perpendicular magnetic anisotropy. We find that the magnon current with spin angular momentum can traverse a 25-nm-thick antiferromagnetic NiO layer and effectively switch the perpendicular magnetization of Co−Fe−B at room temperature with a critical switching current density of 4.1 × 106A/cm2. The magnon torque efficiency is characterized using spin-torque ferromagnetic resonance measurements to be 0.33 with a magnon diffusion length of 26.6 nm. Our work paves the way for manipulating perpendicular magnetization via magnon torques, facilitating the exploration of magnon-based spintronics with low power consumption.

关键词： Magnetic anisotropy Magnetization switching Magnons Spin-orbit torque Ferromagnets Magnetic insulators Magnetic multilayers Topological insulators

来源：评论

学校读者我要写书评

暂无评论

Codec Data Augmentation for Time-Domain Heart Sound Classification

Codec Data Augmentation for Time-Domain Heart Sound Classifi...

引用

International Conference of Advanced Informatics: Concept, Theory and Application (ICAICTA)

作者： Ansh Mishra Jia Qi Yip Eng Siong Chng Department of Physics IIT Delhi New Delhi India Interdisciplinary Graduate Programme Nanyang Technological University (NTU) Singapore School of Computer Science and Engineering Nanyang Technological University (NTU) Singapore

Heart auscultations are a low-cost and effective way of detecting valvular heart diseases early, which can save lives. Nevertheless, it has been difficult to scale this screening method since the effectiveness of auscultations is dependent on the skill of doctors. As such, there has been increasing research interest in the automatic classification of heart sounds using deep learning algorithms. However, it is currently difficult to develop good heart sound classification models due to the limited data available for training. In this work, we propose a simple time domain approach, to the heart sound classification problem with a base classification error rate of 0.8 and show that augmentation of the data through codec simulation can improve the classification error rate to 0.2. With data augmentation, our approach outperforms the existing time-domain CNN-BiLSTM baseline model. Critically, our experiments show that codec data augmentation is effective in getting around the data limitation.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Robust triboelectric information-mat enhanced by multi-modality deep learning for smart home

引用

InfoMat 2023年第1期5卷 139-160页

作者： Yanqin Yang Qiongfeng Shi Zixuan Zhang Xuechuan Shan Budiman Salam Chengkuo Lee Department of Electrical and Computer Engineering National University of SingaporeSingaporeSingapore Center for Intelligent Sensors and MEMS(CISM) National University of SingaporeSingaporeSingapore National University of Singapore Suzhou Research Institute(NUSRI) SuzhouPeople's Republic of China Singapore Institute of Manufacturing Technology and National University of Singapore(SIMTech-NUS)Joint Lab on Large-area Flexible Hybrid Electronics National University of SingaporeSingaporeSingapore Printed Intelligent Device Group Singapore Institute of Manufacturing TechnologyAgency for ScienceTechnology and Research(A*STAR)SingaporeSingapore NUS Graduate School-Integrative Sciences and Engineering Programme(ISEP) National University of SingaporeSingaporeSingapore

In metaverse,a digital-twin smart home is a vital platform for immersive communication between the physical and virtual *** nanogenerators(TENGs)sensors contribute substantially to providing smart-home ***,TENG deployment is hindered by its unstable out-put under environment ***,we develop a digital-twin smart home using a robust all-TENG based information mat(InfoMat),which consists of an in-home mat array and an entry *** interdigital electrodes design allows environment-insensitive ratiometric readout from the mat array to can-cel the commonly experienced environmental *** position sensing is also achieved because of the interval arrangement of the mat ***,the two-channel entry mat generates multi-modality informa-tion to aid the 10-user identification accuracy to increase from 93% to 99% compared to the one-channel ***,a digital-twin smart home is visualized by real-time projecting the information in smart home to virtual reality,including access authorization,position,walking trajectory,dynamic activities/sports,and so on.

关键词： digital twin environment-insensitive multi-modality deep learning scalability smart home triboelectric information-mat

来源：评论

学校读者我要写书评

暂无评论

Resource Allocation of Federated Learning for the Metaverse with Mobile Augmented Reality

arXiv

引用

arXiv 2022年

作者： Zhou, Xinyu Liu, Chang Zhao, Jun The School of Computer Science and Engineering Nanyang Technological University Singapore ERI@N Interdisciplinary Graduate Programme Nanyang Technological University Singapore

The Metaverse has received much attention recently. Metaverse applications via mobile augmented reality (MAR) require rapid and accurate object detection to mix digital data with the real world. Federated learning (FL) is an intriguing distributed machine learning approach due to its privacy-preserving characteristics. Due to privacy concerns and the limited computation resources on mobile devices, we incorporate FL into MAR systems of the Metaverse to train a model cooperatively. Besides, to balance the trade-off between energy, execution latency and model accuracy, thereby accommodating different demands and application scenarios, we formulate an optimization problem to minimize a weighted combination of total energy consumption, completion time and model accuracy. Through decomposing the non-convex optimization problem into two subproblems, we devise a resource allocation algorithm to determine the bandwidth allocation, transmission power, CPU frequency and video frame resolution for each participating device. We further present the convergence analysis and computational complexity of the proposed algorithm. Numerical results show that our proposed algorithm has better performance (in terms of energy consumption, completion time and model accuracy) under different weight parameters compared to existing benchmarks. © 2022, CC BY.

关键词： Frequency division multiple access

来源：评论

学校读者我要写书评

暂无评论

Unifying and Boosting Gradient-Based Training-Free Neural Architecture Search

arXiv

引用

arXiv 2022年

作者： Shu, Yao Dai, Zhongxiang Wu, Zhaoxuan Low, Bryan Kian Hsiang Dept. of Computer Science National University of Singapore Singapore Institute of Data Science National University of Singapore Singapore Integrative Sciences and Engineering Programme NUSGS Singapore

Neural architecture search (NAS) has gained immense popularity owing to its ability to automate neural architecture design. A number of training-free metrics are recently proposed to realize NAS without training, hence making NAS more scalable. Despite their competitive empirical performances, a unified theoretical understanding of these training-free metrics is lacking. As a consequence, (a) the relationships among these metrics are unclear, (b) there is no theoretical interpretation for their empirical performances, and (c) there may exist untapped potential in existing training-free NAS, which probably can be unveiled through a unified theoretical understanding. To this end, this paper presents a unified theoretical analysis of gradient-based training-free NAS, which allows us to (a) theoretically study their relationships, (b) theoretically guarantee their generalization performances, and (c) exploit our unified theoretical understanding to develop a novel framework named hybrid NAS (HNAS) which consistently boosts training-free NAS in a principled way. Remarkably, HNAS can enjoy the advantages of both training-free (i.e., the superior search efficiency) and training-based (i.e., the remarkable search effectiveness) NAS, which we have demonstrated through extensive experiments. Copyright © 2022, The Authors. All rights reserved.

关键词： Machine learning

来源：评论

学校读者我要写书评

暂无评论

Use Your INSTINCT: INSTruction optimization for LLMs usIng Neural bandits Coupled with Transformers

arXiv

引用

arXiv 2023年

作者： Lin, Xiaoqiang Wu, Zhaoxuan Dai, Zhongxiang Hu, Wenyang Shu, Yao Ng, See-Kiong Jaillet, Patrick Low, Bryan Kian Hsiang Department of Computer Science National University of Singapore Singapore Institute of Data Science National University of Singapore Singapore Integrative Sciences and Engineering Programme National University of Singapore Singapore LIDS and EECS Massachusetts Institute of Technology United States China

Large language models (LLMs) have shown remarkable instruction-following capabilities and achieved impressive performances in various applications. However, the performances of LLMs depend heavily on the instructions given to them, which are typically manually tuned with substantial human efforts. Recent work has used the query-efficient Bayesian optimization (BO) algorithm to automatically optimize the instructions given to black-box LLMs. However, BO usually falls short when optimizing highly sophisticated (e.g., high-dimensional) objective functions, such as the functions mapping an instruction to the performance of an LLM. This is mainly due to the limited expressive power of the Gaussian process (GP) which is used by BO as a surrogate to model the objective function. Meanwhile, it has been repeatedly shown that neural networks (NNs), especially pre-trained transformers, possess strong expressive power and can model highly complex functions. So, we adopt a neural bandit algorithm which replaces the GP in BO by an NN surrogate to optimize instructions for black-box LLMs. More importantly, the neural bandit algorithm allows us to naturally couple the NN surrogate with the hidden representation learned by a pre-trained transformer (i.e., an open-source LLM), which significantly boosts its performance. These motivate us to propose our INSTruction optimization usIng Neural bandits Coupled with Transformers (INSTINCT) algorithm. We perform instruction optimization for ChatGPT and use extensive experiments to show that INSTINCT consistently outperforms baselines in different tasks, e.g., various instruction induction tasks and the task of improving zero-shot chain-of-thought instructions. Our code is available at https: //***/xqlin98/INSTINCT. Copyright © 2023, The Authors. All rights reserved.

关键词： Zero-shot learning

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：