检索结果-内蒙古大学图书馆

Brain-inspired artificial intelligence research: A review

Science China(Technological Sciences) 2024年第8期67卷 2282-2296页

作者： WANG GuoYin BAO HuaNan LIU Qun ZHOU TianGang WU Si HUANG TieJun YU ZhaoFei LU CeWu GONG YiHong ZHANG ZhaoXiang HE Sheng Chongqing Key Laboratory of Computational Intelligence Chongqing University of Posts and TelecommunicationsChongqing 400065China Key Laboratory of Cyberspace Big Data Intelligent Security Chongqing University of Posts and TelecommunicationsChongqing 400065China College of Computer and Information Science Chongqing Normal UniversityChongqing 401331China State Key Laboratory of Brain and Cognitive Science Institute of BiophysicsChinese Academy of SciencesBeijing 100101China School of Psychological and Cognitive Sciences Peking UniversityBeijing 100871China State Key Laboratory of Multimedia Information Processing School of Computer SciencePeking UniversityBeijing 100871China Department of Computer Science School of ElectronicsInformation and Electrical EngineeringShanghai Jiao Tong UniversityShanghai 200240China Faculty of Electronic and Information Engineering Xi’an Jiaotong UniversityXi’an 710049China The Center for Research on Intelligent Perception and Computing Institute of AutomationChinese Academy of SciencesBeijing 100190China Institute of Biophysics Chinese Academy of SciencesBeijing 100101China

Artificial intelligence(AI) systems surpass certain human intelligence abilities in a statistical sense as a whole, but are not yet the true realization of these human intelligence abilities and behaviors. There are differences, and even contradictions, between the cognition and behavior of AI systems and humans. With the goal of achieving general AI, this study contains a review of the role of cognitive science in inspiring the development of the three mainstream academic branches of AI based on the three-layer framework proposed by David Marr, and the limitations of the current development of AI are explored and analyzed. The differences and inconsistencies between the cognition mechanisms of the human brain and the computation mechanisms of AI systems are analyzed. They are found to be the cause of the differences and contradictions between the cognition and behavior of AI systems and humans. Additionally, eight important research directions and their scientific issues that need to focus on braininspired AI research are proposed: highly imitated bionic information processing, a large-scale deep learning model that balances structure and function, multi-granularity joint problem solving bidirectionally driven by data and knowledge, AI models that simulate specific brain structures, a collaborative processing mechanism with the physical separation of perceptual processing and interpretive analysis, embodied intelligence that integrates the brain cognitive mechanism and AI computation mechanisms,intelligence simulation from individual intelligence to group intelligence(social intelligence), and AI-assisted brain cognitive intelligence.

关键词： artificial intelligence cognitive science brain science intelligence science large language model

来源：评论

学校读者我要写书评

暂无评论

Multi-Level Speaker Representation for Target Speaker Extraction

Multi-Level Speaker Representation for Target Speaker Extrac...

引用

2025 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2025

作者： Zhang, Ke Li, Junjie Wang, Shuai Wei, Yangjie Wang, Yi Wang, Yannan Li, Haizhou Key Laboratory of Intelligent Computing in Medical Image Northeastern University China SDS SRIBD The Chinese University of Hong Kong Shenzhen China The Hong Kong Polytechnic University Hong Kong Tencent Ethereal Audio Lab Tencent Shenzhen China University of Bremen Germany Department of Electrical and Computer Engineering National University of Singapore Singapore

ISBN: (纸本)9798350368741

Target speaker extraction (TSE) relies on a reference cue of the target to extract the target speech from a speech mixture. While a speaker embedding is commonly used as the reference cue, such embedding pre-trained with a large number of speakers may suffer from confusion of speaker identity. In this work, we propose a multi-level speaker representation approach, from raw features to neural embeddings, to serve as the speaker reference cue. We generate a spectral-level representation from the enrollment magnitude spectrogram as a raw, low-level feature, which significantly improves the model's generalization capability. Additionally, we propose a contextual embedding feature based on cross-attention mechanisms that integrate frame-level embeddings from a pre-trained speaker encoder. By incorporating speaker features across multiple levels, we significantly enhance the performance of the TSE model. Our approach achieves a 2.74 dB improvement and a 4.94% increase in extraction accuracy on Libri2mix test set over the baseline. © 2025 IEEE.

关键词： Cocktail party problem selective auditory attention speaker confusion speaker feature target speaker extraction

来源：评论

学校读者我要写书评

暂无评论

Contactless Interaction System Based on Facial Expression Recognition for Humanoid Piano Robot

Contactless Interaction System Based on Facial Expression Re...

引用

2022 IEEE International Conference on Robotics and Biomimetics, ROBIO 2022

作者： Zhong, Ling Wang, Wen Zhu, Shiqiang Ji, Shanshan Zha, Changhai Wan, Minhong Gu, Jason Research Center for Intelligent Robotics Zhejiang Engineering Research Center for Intelligent Robotics Research Institute of Interdisciplinary Innovation Zhejiang Lab Hangzhou311100 China Dalhousie University Department of Electrical and Computer Engineering HalifaxNSB3H 4R2 Canada

ISBN: (纸本)9781665481090

With the wide application of service robots in people's daily life, people are not only satisfied with robots accomplishing tasks independently but also hope that robots can maintain sustainable interaction with human beings. Especially in the field of music, robots need to be equipped with intelligent cognitive skills in decision-making, so that the audience can enjoy immersive appreciation and emotional resonance in the robots performance. Visual perception and natural language understanding are essential for robots to establish human-robot friendly relationships. To enrich the interactive ability of music robots, we design a contactless interaction system based on facial expression recognition, which consists of four modules: voice wake-up, face detection, facial expression recognition, and music mapping. When people wake up the music robot with specific words, the robot will customize appropriate music to perform according to the recognition results of the FER module. As a final step, the interaction system has been successfully applied to the piano humanoid robot independently developed by our team. © 2022 IEEE.

关键词： Decision making

来源：评论

学校读者我要写书评

暂无评论

Smart and Advanced electrical Skills: Identifying Current Gaps and Emerging Trends

Smart and Advanced Electrical Skills: Identifying Current Ga...

引用

Communications, Information, Electronic and Energy Systems (CIEES), International Conference on

作者： Emmanuel Karapidakis Marini Markaki Spyridon Apostolakis Bruno Canizes Radamanthys Androulakis Fábio Castro Electrical and Computer Engineering Department Hellenic Mediterranean University Heraklion Greece GECAD - Research Group on Intelligent Engineering and Computing for Advanced Innovation and Development Polytechnic Institute of Porto Porto Portugal GECAD - Research Group on Research Department ECTE Rethymno Greece

ISBN: (数字)9798350352863

ISBN: (纸本)9798350352870

With the rise of smart buildings, new expertise gaps and security vulnerabilities are emerging, necessitating a comprehensive understanding of their current state and proactive future planning to address these challenges. Such identification is crucial for mapping the pathway towards vocational modernization (keeping up with trends and goal alignment). To assess these gaps, this paper explores the extensive literature that draws on workshops, expert interviews, and research to identify emerging roles and training needs across seven countries: Greece, Spain, Portugal, Italy, North Macedonia, Germany, and the Netherlands. In particular, it examines the validity of its findings in critical areas such as security, privacy, and the integration of smart grids with energy markets. Additionally, the paper highlights future skills gaps emerging from advancements in IoT, automation, and renewable energy technologies. It emphasizes the growing demand for specialized electrical expertise to optimize energy production, integrate monitoring and storage systems, and manage electric vehicle charging and energy communities effectively. In conclusion, the study identifies critical gaps in expertise and training needed to keep up with the rapid evolution of smart building technologies. Emerging trends highlight the demand for interdisciplinary skills, particularly in IoT, automation, and renewable energy. As these fields advance, expertise in cybersecurity, energy management, and smart grid integration will be crucial, emphasizing the need for proactive workforce development and planning. The findings can be used to inform and direct policymakers towards the smooth reexamination of vocational training.

关键词： Renewable energy sources Smart buildings Automation Production Market research Vocational training Smart grids Planning computer security Energy management

来源：评论

学校读者我要写书评

暂无评论

Exploring Enhancements to the Stacked LSTM Outlier Detection and Correction Method

Exploring Enhancements to the Stacked LSTM Outlier Detection...

引用

International Conference on computer and Applications (ICCA)

作者： Juan Zuluaga Navid Shaghaghi Department of Computer Science and Engineering (CSEN) Ethical Pragmatic and Intelligent Computing (EPIC) Research Lab School of Engineering (SoE) Santa Clara University (SCU) Santa Clara California USA

ISBN: (数字)9798350367560

ISBN: (纸本)9798350367577

Existence of outliers in a dataset render it problematic for Time Series Forecasting. Hence various mathematical and statistical methodologies such as Omission and imputation, Z-Score Normalization, Winsorization, and Robust Regression, have been utilized but often fall short especially when working with datasets containing extensive outlier zones. A novel solution proposed and evaluated by the authors in a previous paper has shown promise especially when utilizing Long Short-Term Memory (LSTM)s, but further enhancements in the form of increasing training data size and variation, ensemble strategies, autoencoders, and smoothing techniques were suggested. This paper reports on the exploration of the effects of these enhancements on said solution.

关键词： Smoothing methods Autoencoders Time series analysis Training data Predictive models Data models Complexity theory Forecasting Long short term memory Anomaly detection

来源：评论

学校读者我要写书评

暂无评论

Purification Of Contaminated Convolutional Neural Networks Via Robust Recovery: An Approach with Theoretical Guarantee in One-Hidden-Layer Case

arXiv

引用

arXiv 2024年

作者： Lu, Hanxiao Huang, Zeyu Wang, Ren The Department of Electrical and Computer Engineering Illinois Institute of Technology ChicagoIL60616 United States The Trustworthy and Intelligent Machine Learning Research Lab The Department of Electrical and Computer Engineering Illinois Institute of Technology ChicagoIL60616 United States

Convolutional neural networks (CNNs), one of the key architectures of deep learning models, have achieved superior performance on many machine learning tasks such as image classification, video recognition, and power systems. Despite their success, CNNs can be easily contaminated by natural noises and artificially injected noises such as backdoor attacks. In this paper, we propose a robust recovery method to remove the noise from the potentially contaminated CNNs and provide an exact recovery guarantee on one-hidden-layer non-overlapping CNNs with the rectified linear unit (ReLU) activation function. Our theoretical results show that both CNNs' weights and biases can be exactly recovered under the overparameterization setting with some mild assumptions. The experimental results demonstrate the correctness of the proofs and the effectiveness of the method in both the synthetic environment and the practical neural network setting. Our results also indicate that the proposed method can be extended to multiple-layer CNNs and potentially serve as a defense strategy against backdoor attacks. Copyright © 2024, The Authors. All rights reserved.

关键词： Recovery

来源：评论

学校读者我要写书评

暂无评论

Crashworthiness Design Optimization for an eVTOL Aircraft

Crashworthiness Design Optimization for an eVTOL Aircraft

引用

2022 IEEE International Conference on Robotics and Biomimetics, ROBIO 2022

作者： Ding, Menglong Xie, Anhuan Zhu, Shiqiang Song, Wei Cai, Jiandong Yan, Xufei Zhao, Pengyu Gu, Jason Zhang, Yanyan Zhejiang Lab Research Center for Intelligent Robotics Research Institute of Interdisciplinary Innovation Hangzhou311100 China Zhejiang Engineering Research Center for Intelligent Robotics Hangzhou311100 China Dalhousie University Department of Electrical and Computer Engineering HalifaxNSB3H4R2 Canada

ISBN: (纸本)9781665481090

The emergence of electric Vertical Takeoff and Landing (eVTOL) aircraft makes cheap, quiet and reliable short-range flights possible. As a manned aircraft, eVTOL's occupant protection capability is critical and its crashworthiness must be improved via optimization of energy absorption design for structures. This study aims to evaluate the crashworthiness of composite components and the entire eVTOL structure within the design space via the building-block approach. A variety of designs for the skid landing gear, energy absorber, structures in the cockpit and the combination of them were estimated by using the numerical method. The results show a tremendous reduction in occupant injury risk with the optimization of the skid landing gear, energy absorber and the entire eVTOL structure. This work provides a preliminary optimization design and evaluation methodology of crashworthiness for an eVTOL. © 2022 IEEE.

关键词： Crashworthiness

来源：评论

学校读者我要写书评

暂无评论

Aerodynamic Simulation of a Typical Multirotor e VTOL Aircraft Configuration

Aerodynamic Simulation of a Typical Multirotor e VTOL Aircra...

引用

2022 IEEE International Conference on Robotics and Biomimetics, ROBIO 2022

作者： Ding, Menglong Zhang, Yanyan Zhu, Shiqiang Song, Wei Zhao, Pengyu Yan, Xufei Cai, Jiandong Gu, Jason Xie, Anhuan Research Institute of Interdisciplinary Innovation Zhejiang Lab Research Center for Intelligent Robotics Hangzhou311100 China Zhejiang Engineering Research Center for Intelligent Robotics Hangzhou311100 China Dalhousie University Department of Electrical and Computer Engineering HalifaxNSB3H4R2 Canada

ISBN: (纸本)9781665481090

The aerodynamic interaction among rotors is a key phenomenon that influences the performance of the electric Vertical Takeoff and Landing (eVTOL) aircraft. This study applied Lattice-Boltzmann Method (LBM) to systematically investigate the aerodynamic interaction under typical flight conditions. Simulation of a single rotor was performed to validate the reliability of LBM on rotor aerodynamics and find the maximum acceptable grid size. A numerical study on the complex unsteady flowfield caused by the sixteen rotors was also performed when considering the eVTOL is in hovering, forward flight and hovering under sidewind. Moreover, the analysis of the flowfield provides a detailed insight into the flow physics involved in such regular flight conditions. © 2022 IEEE.

关键词： VTOL/STOL aircraft

来源：评论

学校读者我要写书评

暂无评论

New In-Air Signature Datasets

New In-Air Signature Datasets

引用

2023 International Symposium on Networks, computers and Communications, ISNCC 2023

作者： Alobaidi, Abdulazeez R. Dhieb, Thameur Nuimi, Zeina N. Hamdani, Tarek M. Wali, Ali Alimi, Adel M. University of Sfax REGIM-Lab.: REsearch Groups in Intelligent Machines Enis Tunisia Imam ja'Afar Al-Sadiq University Faculty of Information Technology Department of Computer Techniques Engineering Iraq Higher Institute of Business Administration University of Gafsa Tunisia College of Engineering Department of Electrical Engineering University of Misan Iraq Higher Institute of Computer Science Mahdia University of Monastir Tunisia Higher Institute of Computer Science and Multimedia of Sfax University of Sfax Tunisia University of Johannesburg Faculty of Engineering and the Built Environment Department of Electrical and Electronic Engineering Science South Africa

ISBN: (纸本)9798350335590

Compared to traditional biometric systems, in-air signatures are considered more robust and secure than classical pen paper. A few datasets capturing in-air signatures have been introduced, utilizing various devices such as the Leap Motion and the Microsoft Kinect sensor camera. However, these devices are not exempt from shortcomings and exhibit certain limitations. The expenses associated with their implementation and the requirement for technical proficiency in operating them present notable challenges for in-air signature analysis. Additionally, users may encounter difficulties in adapting their finger movements to fit within the device's limited field of view, particularly if they lack familiarity with these devices. To address these concerns, this paper proposes the creation of three in-air signature datasets using solely the camera of a laptop or a smartphone, eliminating the need for any additional specialized equipment. Our datasets were collected in three ways. The first is the In-Air Signature dataset (IAS dataset) and the second is the In-Air Signature dataset using a transparent Glass Plate (IASGP dataset) while the third is the In-Air Signature dataset using Smart Phone (IASSP dataset). Forty volunteers participated in the construction of these datasets. Their ages ranged from 21 to 40 years. Each volunteer signs in the air five signatures and imitates five signatures of five other volunteers. Our in-air signatures datasets are publicly available and can be used for various research tasks like in-air signature verification and identification. © 2023 IEEE.

关键词： Cameras

来源：评论

学校读者我要写书评

暂无评论

Advancing Proactive Producer Mobility Management in Named Data Networking: A Conceptual Model 7

Advancing Proactive Producer Mobility Management in Named Da...

引用

7th International Conference on Internet Applications, Protocols, and Services, NETAPPS 2024

作者： Zukri, Nurul Hidayah Ahmad Arif, Ahmad Suki Che Mohamed Alsamman, Mohammed Abrar, Ahmad Fazea, Yousef Universiti Teknologi MARA Cawangan Perlis Kampus Arau Perlis Malaysia School of Computing Universiti Utara Malaysia Kedah Malaysia InterNetWorks Research Lab Universiti Utara Malaysia Kedah Malaysia Department of Computer Science and Electrical Engineering Marshall University One John Marshall Dr. HuntingtonWV25755 United States

ISBN: (纸本)9798331519513

In recent years, the Named Data Networking (NDN) paradigm has emerged as a promising candidate for future Internet architecture. Underpinned by the notion of a data-centric forwarding philosophy, NDN is purported to offer seamless network communication for mobile users. However, while it provides inherent support for consumer mobility, viable solutions for producer mobility remain insufficient. Therefore, this study aims to highlight the mobility issues in NDN, particularly concerning high latency, Interest packet loss, and Interest retransmission, with a focus on producer mobility. In this paper, we critically review previous works based on proactive and reactive approaches. Furthermore, we propose a mobility management model that adopts a proactive approach, estimating the producer's handoff time from the connected access point. The proposed model leverages this information to notify consumers about the moving producer, aiming to prevent consumers from sending Interest to the mobile producer during handoff and thereby overcome the associated challenges. © 2024 IEEE.

关键词： Named Data Network Proactive Producer Mobility

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：