检索结果-内蒙古大学图书馆

arXiv 2021年

作者： Han, Zongyan Fu, Zhenyong Chen, Shuo Yang, Jian PCALab Nanjing University of Science and Technology China Riken Center for Advanced Intelligence Project Japan PCA Lab Key Lab of Intelligent Percept. and Syst. for High-Dimensional Information of Ministry of Education Jiangsu Key Lab of Image and Video Understanding for Social Security School of Computer Science and Engineering Nanjing University of Science and Technology China

Generalized zero-shot learning (GZSL) aims to recognize objects from both seen and unseen classes, when only the labeled examples from seen classes are provided. Recent feature generation methods learn a generative model that can synthesize the missing visual features of unseen classes to mitigate the data-imbalance problem in GZSL. However, the original visual feature space is suboptimal for GZSL classification since it lacks discriminative information. To tackle this issue, we propose to integrate the generation model with the embedding model, yielding a hybrid GZSL framework. The hybrid GZSL approach maps both the real and the synthetic samples produced by the generation model into an embedding space, where we perform the final GZSL classification. Specifically, we propose a contrastive embedding (CE) for our hybrid GZSL framework. The proposed contrastive embedding can leverage not only the class-wise supervision but also the instance-wise supervision, where the latter is usually neglected by existing GZSL researches. We evaluate our proposed hybrid GZSL framework with contrastive embedding, named CE-GZSL, on five benchmark datasets. The results show that our CEGZSL method can outperform the state-of-the-arts by a significant margin on three datasets. Our codes are available on https://***/Hanzy1996/CE-GZSL. Copyright © 2021, The Authors. All rights reserved.

关键词： Embeddings

来源：评论

学校读者我要写书评

暂无评论

Learning interpretable end-to-end vision-based motion planning for autonomous driving with optical flow distillation

arXiv

引用

arXiv 2021年

作者： Wang, Hengli Cai, Peide Sun, Yuxiang Wang, Lujia Liu, Ming Department of Electronic and Computer Engineering Hong Kong University of Science and Technology Clear Water Bay Kowloon Hong Kong Department of Mechanical Engineering Hong Kong Polytechnic University Hung Hom Kowloon Hong Kong Cloud Computing Lab of Shenzhen Institutes of Advanced Technology Chinese Academy of Sciences China

Recently, deep-learning based approaches have achieved impressive performance for autonomous driving. However, end-to-end vision-based methods typically have limited interpretability, making the behaviors of the deep networks difficult to explain. Hence, their potential applications could be limited in practice. To address this problem, we propose an interpretable end-to-end vision-based motion planning approach for autonomous driving, referred to as IVMP. Given a set of past surrounding-view images, our IVMP first predicts future egocentric semantic maps in bird's-eye-view space, which are then employed to plan trajectories for self-driving vehicles. The predicted future semantic maps not only provide useful interpretable information, but also allow our motion planning module to handle objects with low probability, thus improving the safety of autonomous driving. Moreover, we also develop an optical flow distillation paradigm, which can effectively enhance the network while still maintaining its real-time performance. Extensive experiments on the nuScenes dataset and closed-loop simulation show that our IVMP significantly outperforms the state-of-the-art approaches in imitating human drivers with a much higher success rate. Our project page is available at https://***/view/ivmp. Copyright © 2021, The Authors. All rights reserved.

关键词： Autonomous vehicles

来源：评论

学校读者我要写书评

暂无评论

Social Choice Should Guide AI Alignment in Dealing with Diverse Human Feedback

arXiv

引用

arXiv 2024年

作者： Conitzer, Vincent Freedman, Rachel Heitzig, Jobst Holliday, Wesley H. Jacobs, Bob M. Lambert, Nathan Mossé, Milan Pacuit, Eric Russell, Stuart Schoelkopf, Hailey Tewolde, Emanuel Zwicker, William S. Foundations of Cooperative AI Lab Computer Science Department Carnegie Mellon University Pittsburgh United States Institute for Ethics in AI University of Oxford Oxford United Kingdom Center for Human-Compatible AI Department of Electrical Engineering and Computer Sciences University of California Berkeley United States Potsdam Institute for Climate Impact Research Brandenburg Potsdam Germany Department of Philosophy University of California Berkeley United States Department of Philosophy and Moral Sciences Ghent University Ghent Belgium Allen Institute for AI BerkeleyCA United States Department of Philosophy University of Maryland College Park United States EleutherAI Greece Department of Mathematics Union College Schenectady United States Murat Sertel Center for Advanced Economic Studies Istanbul Bilgi University Istanbul Turkey

Foundation models such as GPT-4 are fine-tuned to avoid unsafe or otherwise problematic behavior, such as helping to commit crimes or producing racist text. One approach to fine-tuning, called reinforcement learning from human feedback, learns from humans’ expressed preferences over multiple outputs. Another approach is constitutional AI, in which the input from humans is a list of high-level principles. But how do we deal with potentially diverging input from humans? How can we aggregate the input into consistent data about "collective" preferences or otherwise use it to make collective choices about model behavior? In this paper, we argue that the field of social choice is well positioned to address these questions, and we discuss ways forward for this agenda, drawing on discussions in a recent workshop on Social Choice for AI Ethics and Safety held in Berkeley, CA, USA in December *** Codes 68T01, 68T50, 91B14, 91B12 Copyright © 2024, The Authors. All rights reserved.

关键词： Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

A Human-Machine Joint Learning Framework to Boost Endogenous BCI Training

arXiv

引用

arXiv 2023年

作者： Wang, Hanwen Qi, Yu Yao, Lin Wang, Yueming Farina, Dario Pan, Gang The College of Computer Science and Technology Zhejiang University Hangzhou China The Affiliated Mental Health Center Hangzhou Seventh People's Hospital The MOE Frontier Science Center for Brain Science and Brain-machine Integration Zhejiang University School of Medicine The State Key Lab of Brain-Machine Intelligence Hangzhou China The Department of Neurobiology Affiliated Mental Health Center Hangzhou Seventh People's Hospital Zhejiang University School of Medicine MOE Frontiers Science Center for Brain and Brain-Machine Integration Zhejiang University Department of Biomedical Engineering Zhejiang University College of Computer Science and Technology Hangzhou China The Qiushi Academy for Advanced Studies Zhejiang University Hangzhou China The Department of Bioengineering Imperial College London London United Kingdom The State Key Lab of Brain-Machine Intelligence The College of Computer Science and Technology Zhejiang University Hangzhou China The First Affiliated Hospital Zhejiang University Hangzhou China

Brain-computer interfaces (BCIs) provide a direct pathway from the brain to external devices and have demonstrated great potential for assistive and rehabilitation technologies. Endogenous BCIs based on electroencephalogram (EEG) signals, such as motor imagery (MI) BCIs, can provide some level of control. However, mastering spontaneous BCI control requires the users to generate discriminative and stable brain signal patterns by imagery, which is challenging and is usually achieved over a very long training time (weeks/months). Here, we propose a human-machine joint learning framework to boost the learning process in endogenous BCIs, by guiding the user to generate brain signals towards an optimal distribution estimated by the decoder, given the historical brain signals of the user. To this end, we firstly model the human-machine joint learning process in a uniform formulation. Then a human-machine joint learning framework is proposed: 1) for the human side, we model the learning process in a sequential trial-and-error scenario and propose a novel "copy/new" feedback paradigm to help shape the signal generation of the subject toward the optimal distribution;2) for the machine side, we propose a novel adaptive learning algorithm to learn an optimal signal distribution along with the subject's learning process. Specifically, the decoder reweighs the brain signals generated by the subject to focus more on "good" samples to cope with the learning process of the subject. Online and psuedo-online BCI experiments with 18 healthy subjects demonstrated the advantages of the proposed joint learning process over co-adaptive approaches in both learning efficiency and effectiveness. © 2023, CC BY.

关键词： Electroencephalography

来源：评论

学校读者我要写书评

暂无评论

Threat-Based Declassification and Endorsement for Mobile Computing

引用

Chinese Journal of Electronics 2019年第5期28卷 1041-1052页

作者： YIN Lihua GUO Yunchuan ZHANG Huibing HUANG Wenting FANG Binxing Cyberspace Institute of Advanced Technology Guangzhou University The State Key Laboratory of Information Security Institute of Information Engineering CAS Guangxi Key Lab of Trusted Software Guilin University of Electronic Technology The National Computer Network Emergency Response Technical Team/Coordination Center of China

Declassification and endorsement can efficiently improve the usability of mobile ***, both declassify and endorse operations in practice are often ad-hoc and nondeterministic, thus, being insecure. From a new perspective of threat assessments,we propose the Threat-based typed security π-calculus(πTBTS) to model declassification and endorsement in mobile computing. Intuitively, when relaxing confidentiality policies and/or integrity policies, we respectively assess threats brought by performing these two relaxes. If these threats are acceptable, the declassification and/or endorsement operations are permitted; Otherwise, they are denied. The proposed assessments have explicit security conditions, results and less open parameters,so our approach solves the problem of the ad-hoc and nondeterministic semantics and builds a bridge between threat assessments and declassification/endorsement.

关键词： Declassification Endorsement Threat assessment Mobile computing π-Calculus

来源：评论

学校读者我要写书评

暂无评论

Cardiovascular disease identification using a hybrid CNN-LSTM model with explainable AI

引用

Informatics in Medicine Unlocked 2023年 42卷

作者： Hossain, Md Maruf Ali, Md Shahin Ahmed, Md Mahfuz Rakib, Md Rakibul Hasan Kona, Moutushi Akter Afrin, Sadia Islam, Md Khairul Ahsan, Md Manjurul Raj, Sheikh Md Razibul Hasan Rahman, Md Habibur Department of Biomedical Engineering Islamic University Kushtia 7003 Bangladesh Department of Food and Nutrition Government College of Applied Human Science Azimpur Dhaka 1205 Bangladesh Department of Radiology Northwestern University Machine and Hybrid Intelligence Lab Chicago 60611 IL United States Industrial & Production Engineering Shahjalal University of Science and Technology Sylhet 3114 Bangladesh Department of Information and Communication Technology Islamic University Kushtia 7003 Bangladesh Department of Computer Science and Engineering Islamic University Kushtia 7003 Bangladesh Center for Advanced Bioinformatics and Artificial Intelligent Research Islamic University Kushtia 7003 Bangladesh

Cardiovascular disease (CVD) is a leading cause of death worldwide, with millions dying each year. The identification and early diagnosis of CVD are critical in preventing adverse health outcomes. Hence, this study proposes a hybrid deep learning (DL) model that combines a convolutional neural network (CNN) and long short-term memory (LSTM) to identify CVD from the clinical data. This study utilizes CNN to extract the relevant features from the input data and the LSTM network to process sequential data and capture dependencies and patterns over time. This study provides insights into the potential of a hybrid DL model combined with feature engineering and explainable AI to improve the accuracy and interpretability of CVD prediction. We evaluated our model on a publicly available dataset where the proposed CNN-LSTM achieved a high accuracy of 73.52% and 74.15% with and without feature engineering, respectively, in identifying individuals with CVD, which is the best result compared to the current state-of-the-art model. The results of this study demonstrate the potential of DL models for the early diagnosis of CVD. Our proposed CNN-LSTM model also incorporates explainable AI to identify the top features responsible for CVD. They could be used to develop more eﬀective screening tools in clinical practice. © 2023 The Authors

关键词： Cardiovascular disease CNN-LSTM Deep learning Explainable AI Feature engineering

来源：评论

学校读者我要写书评

暂无评论

NAS-Count: Counting-by-Density with Neural Architecture Search 16th

NAS-Count: Counting-by-Density with Neural Architecture Sear...

引用

16th European Conference on computer Vision, ECCV 2020

作者： Hu, Yutao Jiang, Xiaolong Liu, Xuhui Zhang, Baochang Han, Jungong Cao, Xianbin Doermann, David School of Electronic and Information Engineering Beihang University Beijing China Key Laboratory of Advanced Technologies for Near Space Information Systems Ministry of Industry and Information Technology Beijing China Beijing Advanced Innovation Center for Big Data-Based Precision Medicine Beijing China YouKu Cognitive and Intelligent Lab Alibaba Group Hangzhou China Beihang University Beijing China Computer Science Department Aberystwyth University AberystwythSY23 3FL United Kingdom Department of Computer Science and Engineering University at Buffalo New York United States

ISBN: (纸本)9783030585419

Most of the recent advances in crowd counting have evolved from hand-designed density estimation networks, where multi-scale features are leveraged to address the scale variation problem, but at the expense of demanding design efforts. In this work, we automate the design of counting models with Neural Architecture Search (NAS) and introduce an end-to-end searched encoder-decoder architecture, Automatic Multi-Scale Network (AMSNet). Specifically, we utilize a counting-specific two-level search space. The encoder and decoder in AMSNet are composed of different cells discovered from micro-level search, while the multi-path architecture is explored through macro-level search. To solve the pixel-level isolation issue in MSE loss, AMSNet is optimized with an auto-searched Scale Pyramid Pooling Loss (SPPLoss) that supervises the multi-scale structural information. Extensive experiments on four datasets show AMSNet produces state-of-the-art results that outperform hand-designed models, fully demonstrating the efficacy of NAS-Count. © 2020, Springer Nature Switzerland AG.

关键词： Decoding

来源：评论

学校读者我要写书评

暂无评论

Abnormal Data Cleaning for Wind Turbines by Image Segmentation Based on Active Shape Model and Class Uncertainty

SSRN

引用

SSRN 2022年

作者： Liang, Guoyuan Su, Yahao Wu, Xinyu Ma, Jiajun Long, Huan Song, Zhe Guangdong Provincial Key Lab of Robotics and Intelligent System Shenzhen Institute of Advanced Technology Chinese Academy of Sciences Guangdong Province Shenzhen China Guangdong Provincial Key Laboratory of Computer Vision and Virtual Reality Technology Shenzhen Institute of Advanced Technology Chinese Academy of Sciences Guangdong Province Shenzhen China Guangdong-Hong Kong-Macao Joint Laboratory of Human-Machine Intelligence-Synergy Systems Shenzhen Institute of Advanced Technology Chinese Academy of Sciences Guangdong Province Shenzhen China School of Electrical Engineering Southeast University Jiang Su Province Nanjing China School of Business Nanjing University Jiang Su Province Nanjing China

Wind power curve describes the relationship between wind speed and output power of wind turbine, which may be contaminated due to various unexpected factors. Following the idea of image segmentation in our previous work, in this paper, we propose a novel image-based abnormal data cleaning algorithm aiming to provide high quality data for wind power curve modeling and relevant applications. For image binarization, a multiscale space is built to determine the optimal scale at which the most stable as well as informative shape representations are extracted. The Active Shape Model is utilized to model the power curve shape and generate a set of segmentation candidates for filtering abnormal data. To measure the cost of each candidate, the Class Uncertainty map based on dual feature images is built. And eventually the optimal segmentation boundary is determined by the optimization in the energy space. A series of comparative experiments conducted over the real-world datasets of 37 wind turbines have validated the superior performance of the proposed algorithm. © 2022, The Authors. All rights reserved.

关键词： Wind power

来源：评论

学校读者我要写书评

暂无评论

Charge Detrapping on Gate Edge of AlGaN/GaN HEMT Under Drain Stress

Charge Detrapping on Gate Edge of AlGaN/GaN HEMT Under Drain...

引用

International Conference on Solid-State and Integrated Circuit Technology

作者： Rongkang Niu Tianjiao Dai Xinxin Zhang Kuanchang Chang Xinnan Lin The Shenzhen Key Lab of Advanced Electron Device and Integration School of Electronic and computer Engineering Peking University Shenzhen Graduate School Shenzhen China

ISBN: (数字)9781728162355

ISBN: (纸本)9781728162362

This work reports on the mechanism of abnormal dynamic on-resistance (dynamic Ron) in AlGaN/GaN high electron mobility transistors (HEMT) under drain stress (V DS , stress). Based on the capacitance change, it is proposed that the defects in GaN/SiN interface have been considered as the main cause of dynamic Ron decreasing. The time constant of charge detrapping at the edge of the gate (access region near the gate) is greater than 0.05s. It is found that field plate can suppress the unstable factors of dynamic Ron by reducing the peak electric field to suppress electron detrapping.

关键词： Logic gates Stress Wide band gap semiconductors Aluminum gallium nitride HEMTs Gallium nitride Capacitance

来源：评论

学校读者我要写书评

暂无评论

Integration of ALD high-k dipole layers into CMOS SOI nanowire FETs for bi-directional threshold voltage engineering

Integration of ALD high-k dipole layers into CMOS SOI nanowi...

引用

IEEE Silicon Nanoelectronics Workshop (SNW)

作者： Wonil Chung Dongqi Zheng Wei-E Wang Mark Rodder Peide D. Ye School of Electrical and Computer Engineering Purdue University West Lafayette IN USA Samsung Semiconductor Inc Advanced Logic Lab Austin TX USA

ISBN: (数字)9781728197357

ISBN: (纸本)9781728197364

In this report, we successfully demonstrate V T (threshold voltage) tuning in both n- and p- SOI nanowire FETs (NWFET) using ultrathin atomic layer deposition (ALD) dipole layers (Y 2 O 3 and Al 2 O 3 ) inserted directly under the ALD HfO 2 (3 nm). 0.7 nm of Y 2 O 3 inserted between bottom SiO x (2 (3 nm) shifted the V T by -138 mV and -58 mV for n- and p-NWFET, respectively while 0.7 nm Al 2 O 3 shifted the V T of n-NWFET by +219 mV and +134 mV for p-NWFET. V T shift on planar SOI FETs were also observed with the similar trend only when such ALD dipole layers were inserted in direct contact with SiO x . MOS capacitor's V FB (flat band voltage) showed similar tunability with ALD dipole layer positioned near the SiO x /Si channel interface. Intermixing multiple dipole layers can more precisely tune the V T in desired direction and strengths by partially neutralizing each other.

关键词：

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：