检索结果-内蒙古大学图书馆

arXiv 2022年

作者： Yu, Junchi Liang, Jian He, Ran NLPR&CRIPAC Institute of Automation Chinese Academy of Sciences China University of Chinese Academy of Sciences China Center for Excellence in Brain Science and Intelligence Technology CAS China

This paper focuses on out-of-distribution generalization on graphs where performance drops due to the unseen distribution shift. Previous graph domain generalization works always resort to learning an invariant predictor among different source domains. However, they assume sufficient source domains are available during training, posing huge challenges for realistic applications. By contrast, we propose a new graph domain generalization framework, dubbed as DPS, by constructing multiple populations from the source domains. Specifically, DPS aims to discover multiple Diverse and Predictable Subgraphs with a set of generators, namely, subgraphs are different from each other but all the them share the same semantics with the input graph. These generated source domains are exploited to learn an equi-predictive graph neural network (GNN) across domains, which is expected to generalize well to unseen target domains. Generally, DPS is model-agnostic that can be incorporated with various GNN backbones. Extensive experiments on both node-level and graph-level benchmarks shows that the proposed DPS achieves impressive performance for various graph domain generalization tasks. Code is attached in the supplementary. © 2022, CC BY-NC-ND.

关键词： Graph neural networks

来源：评论

学校读者我要写书评

暂无评论

Detecting movement timing from primate intracortical signals for brain-machine interfaces in dynamic environment

Detecting movement timing from primate intracortical signals...

引用

2021 IEEE International Conference on Real-Time Computing and Robotics, RCAR 2021

作者： Li, Chenyang Zhang, Yiheng Wang, Tianwei Xu, Xinxu Wang, Qifan Zheng, Ruichen Cui, He Intelligence Technology and Institute of Neuroscience CAS Key Lab of Primate Neurobiology Center for Excellence in Brain Science Shanghai China University of Chinese Academy of Sciences Beijing100049 China

ISBN: (纸本)9781665436786

Decoding movement parameters, such as direction and speed, has been successfully utilized in brain-machine interfaces (BMIs) for closed-loop controlling an actuator to approach static targets. However, to interact with moving objects demands movement in right direction at right time. Unlike decoding motor variables, detecting movement timing is an outlier detection of recorded intracortical signals from unlabeled dataset. Recent neurophysiological studies have suggested that neuronal activity in motor cortex carries information regarding movement timing, allowing us to design a movement timing detector for voluntary control in open-loop BMI. Here, we have tested the performance of four detectors for movement initiation from recorded neuronal dataset as the first step toward a novel decoding module of BMIs in dynamic environment. © 2021 IEEE.

关键词： Timing circuits

来源：评论

学校读者我要写书评

暂无评论

PIC 4th Challenge: Semantic-Assisted multi-feature encoding and multi-head decoding for dense video captioning 4

PIC 4th Challenge: Semantic-Assisted multi-feature encoding ...

引用

4th Person in Context Workshop, PIC 2022

作者： Lu, Yifan Zhang, Ziqi Chen, Yuxin Yuan, Chunfeng Li, Bing Hu, Weiming National Laboratory of Pattern Recognition CASIA and School of Artificial Intelligence University of Chinese Academy of Sciences Beijing China National Laboratory of Pattern Recognition CASIA Beijing China National Laboratory of Pattern Recognition CASIA and CAS Center for Excellence in Brain Science and Intelligence Technology Beijing China

ISBN: (纸本)9781450394895

The task of Dense Video Captioning (DVC) aims to generate captions with timestamps for multiple events in one video. Semantic information plays an important role for both localization and description of DVC. We present a semantic-Assisted dense video captioning model based on the encoding-decoding framework. In the encoding stage, we design a concept detector to extract semantic information, which is then fused with multi-modal visual features to sufficiently represent the input video. In the decoding stage, we design a classification head, paralleled with the localization and captioning heads, to provide semantic supervision. Our method achieves significant improvements on the YouMakeup dataset \citewang2019youmakeup under DVC evaluation metrics and achieves high performance in the Makeup Dense Video Captioning (MDVC) task of \hrefhttp://***/challenge/task/mdvc/ PIC 4th Challenge. © 2022 ACM.

关键词： Semantics

来源：评论

学校读者我要写书评

暂无评论

Motif-topology and Reward-learning improved Spiking Neural Network for Efficient Multi-sensory Integration

arXiv

引用

arXiv 2022年

作者： Jia, Shuncheng Zuo, Ruichen Zhang, Tielin Liu, Hongxing Xu, Bo Institute of Automation Chinese Academy of Sciences China School of Artificial Intelligence University of Chinese Academy of Sciences China Center for Excellence in Brain Science and Intelligence Technology CAS China School of Information and Electronics Beijing Institute of Technology China

Network architectures and learning principles are key in forming complex functions in artificial neural networks (ANNs) and spiking neural networks (SNNs). SNNs are considered the new-generation artificial networks by incorporating more biological features than ANNs, including dynamic spiking neurons, functionally specified architectures, and efficient learning paradigms. In this paper, we propose a Motif-topology and Reward-learning improved SNN (MR-SNN) for efficient multi-sensory integration. MR-SNN contains 13 types of 3-node Motif topologies which are first extracted from independent single-sensory learning paradigms and then integrated for multi-sensory classification. The experimental results showed higher accuracy and stronger robustness of the proposed MR-SNN than other conventional SNNs without using Motifs. Furthermore, the proposed reward learning paradigm was biologically plausible and can better explain the cognitive McGurk effect caused by incongruent visual and auditory sensory signals. © 2022, CC BY.

关键词： Topology

来源：评论

学校读者我要写书评

暂无评论

ODE-based Recurrent Model-free Reinforcement Learning for POMDPs

arXiv

引用

arXiv 2023年

作者： Zhao, Xuanle Zhang, Duzhen Han, Liyuan Zhang, Tielin Xu, Bo Institute of Automation Chinese Academy of Sciences Beijing China School of Artificial Intelligence University of Chinese Academy of Sciences Beijing China Center for Excellence in Brain Science and Intelligence Technology Chinese Academy of Sciences Shanghai China

Neural ordinary differential equations (ODEs) are widely recognized as the standard for modeling physical mechanisms, which help to perform approximate inference in unknown physical or biological environments. In partially observable (PO) environments, how to infer unseen information from raw observations puzzled the agents. By using a recurrent policy with a compact context, context-based reinforcement learning provides a flexible way to extract unobservable information from historical transitions. To help the agent extract more dynamics-related information, we present a novel ODE-based recurrent model combined with model-free reinforcement learning (RL) framework to solve partially observable Markov decision processes (POMDPs). We experimentally demonstrate the efficacy of our methods across various PO continuous control and meta-RL tasks. Furthermore, our experiments illustrate that our method is robust against irregular observations, owing to the ability of ODEs to model irregularly-sampled time series. Copyright © 2023, The Authors. All rights reserved.

关键词： Ordinary differential equations

来源：评论

学校读者我要写书评

暂无评论

Identification of CHMP4C as a new risk gene for inherited dilated cardiomyopathy

引用

Journal of Genetics and Genomics 2022年第2期49卷 169-172页

作者： Nianwei Zhou Lu Tang Yingying Jiang Xuejie Li Shifang Shan Bo Yuan Shengmei Qin Cuizhen Pan Xiaolin Wang Xianhong Shu Zilong Qiu Junbo Ge Department of Echocardiography Zhongshan HospitalFudan UniversityShanghai Institute of Cardiovascular DiseaseShanghai Institute of Medical ImagingShanghaiChina Institute of Neuroscience State Key Laboratory of NeuroscienceCAS Center for excellence in Brain Science and Intelligence TechnologyChinese Academy of SciencesShanghaiChina Department of Cardiology Zhongshan HospitalFudan UniversityShanghai Institute of Cardiovascular DiseaseShanghaiChina Department of Interventional Radiology Zhongshan HospitalFudan UniversityShanghai Institute of Medical ImagingShanghaiChina

Dilated cardiomyopathy(DCM) is a primary myocardial disease of unknown cause that is characterized by ventricular enlargement and ventricular systolic dysfunction(Reichart et al., 2019). DCM exhibits obvious heterogeneity, and its outcomes extend from arrhythmia to heart failure. Early arrhythmia is a common condition that may progressively become aggravated, and death can occur at any stage of the disease. The most serious complications are heart failure and sudden death(Mc Nally et al., 2013).

关键词： cardiomyopathy ventricular

来源：评论

学校读者我要写书评

暂无评论

Inverse Reinforcement Learning with Attention-based Feature Extraction from Video Demonstrations

Inverse Reinforcement Learning with Attention-based Feature ...

引用

IEEE International Conference on Robotics and Biomimetics

作者： Weixin Zhang Tao Lu Yinghao Cai Shuo Wang State Key Laboratory of Multimodal Artificial Intelligence Systems Institute of Automation Chinese Academy of Sciences Beijing China School of Artificial Intelligence University of Chinese Academy of Sciences Beijing China University of Chinese Academy of Sciences Beijing China CAS Center for Excellence in Brain Science and Intelligence Technology Shanghai China

The sparse reward problem makes the robot difficult to get effective feedback during learning skills with reinforcement learning, which will seriously affect the training efficiency. In this paper, we propose an inverse reinforcement learning (IRL) method to understand expert intent from video demonstrations and further provide dense rewards for robots to learn the task. We learn representations related to task progress by matching different videos in the embedding space based on their temporal alignment and incorporating attention mech-anisms to focus on the task-relevant features. Based on this, we design an efficient and robust dense reward function that enhances the efficiency of the reinforcement learning process. We validated our method in robotic manipulation tasks, and the experimental results showed that our approach exhibits faster convergence speed and stability and have the generalization ability in skill learning with video demonstrations from unseen embodiment.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Inconsistency-Aware Wavelet Dual-Branch Network for Face Forgery Detection

IEEE Transactions on Biometrics, Behavior, and Identity Scie...

引用

IEEE Transactions on Biometrics, Behavior, and Identity science 2021年第3期3卷 308-319页

作者： Jia, Gengyun Zheng, Meisong Hu, Chuanrui Ma, Xin Xu, Yuting Liu, Luoqi Deng, Yafeng He, Ran School of Artificial Intelligence University of Chinese Academy of Sciences Beijing100049 China 360 Ai Institute Beijing Qihu Keji Company Ltd. Beijing China 360 Ai Institute Beijing Qihu Keji Company Ltd. Beijing100020 China National Laboratory of Pattern Recognition Center for Research on Intelligent Perception and Computing Cas Center for Excellence in Brain Science and Intelligence Technology Institute of Automation Chinese Academy of Sciences Beijing100190 China

Current face forgery techniques can generate high-fidelity fake faces with extremely low labor and time costs. As a result, face forgery detection becomes an important research topic to prevent technology abuse. In this paper, we present an inconsistency-aware wavelet dual-branch network for face forgery detection. This model is mainly based on two kinds of forgery clues called inter-image and intra-image inconsistencies. To fully utilize them, we firstly enhance the forgery features by using additional inputs based on stationary wavelet decomposition (SWD). Then, considering the different properties of the two inconsistencies, we design a dual-branch network that predicts image-level and pixel-level forgery labels respectively. The segmentation branch aims to recognize real and fake local regions, which is crucial for discovering intra-image inconsistency. The classification branch learns to discriminate the real and fake images globally, thus can extract inter-image inconsistency. Finally, bilinear pooling is employed to fuse the features from the two branches. We find that the bilinear pooling is a kind of spatial attentive pooling. It effectively utilizes the rich spatial features learned by the segmentation branch. Experimental results show that the proposed method surpasses the state-of-the-art face forgery detection methods. © 2019 IEEE.

关键词： Wavelet decomposition

来源：评论

学校读者我要写书评

暂无评论

HTR2A agonists play a therapeutic role by restricting ILC2 activation in papain-induced lung inflammation

引用

Cellular & Molecular Immunology 2023年第4期20卷 404-418页

作者： Zhishuo Wang Chenghua Yan Qizhen Du Yuying Huang Xuezhen Li Dan Zeng Ruizhi Mao Rama Krishna Gurram Shipeng Cheng Wangpeng Gu Lin Zhu Weiguo Fan Liyan Ma Zhiyang Ling Ju Qiu Dangsheng Li Enmei Liu Yaguang Zhang Yiru Fang Jinfang Zhu Bing Sun state Key Laboratory of Cell Biology CAS Center for Excellence in Molecular Cell ScienceShanghai Institute of Biochemistry and Cell BiologyChinese Academy of SciencesUniversity of Chinese Academy of Sciences320 Yueyang RoadShanghai 200031China College of Life Sciences Jiangxi University of Chinese MedicineNanchang 330004China Department of Respiratory Medicine Children's Hospital of Chongqing Medical University National Clinical Research Center for Child Health and DisordersMinistry of Education Key Laboratory of Child Development and DisordersChongqing Key Laboratory of PediatricsChongqingChina Department of Allergy Chongqing General HospitalChongqingChina Clinical Research Center and Division of Mood Disorders Shanghai Mental Health CenterShanghai Jiao Tong University School of MedicineShanghai 200030China Laboratory of Immune System Biology National Institute of Allergy and Infectious DiseasesNational Institutes of HealthBethesdaMD 20892USA CAS Key Laboratory of Tissue Microenvironment and Tumor Shanghai Institute of Nutrition and HealthUniversity of Chinese Academy of SciencesChinese Academy of SciencesShanghai 200031China CAS Center for Excellence in Brain Science and Intelligence Technology Shanghai 200031China Shanghai Key Laboratory of Psychotic Disorders Shanghai 201108China.

Group 2 innate lymphoid cells(ILC2s)are a category of heterogeneous cells that produce the cytokines IL-5 and IL-13,which mediate the type 2 immune ***,specific drug targets on lung ILC2s have rarely been *** studies have shown that type 2 cytokines,such as IL-5 and IL-13,are related to ***,we demonstrated the negative correlation between the depression-associated monoamine neurotransmitter serotonin and secretion of the cytokines IL-5 and IL-13 by ILC2s in individuals with ***,serotonin ameliorates papain-induced lung inflammation by suppressing ILC2 *** data showed that the serotonin receptor HTR2A was highly expressed on ILC2s from mouse lungs and human ***,an HTR2A selective agonist(DOI)impaired ILC2 activation and alleviated the type 2 immune response in vivo and in *** with ILC2-specific depletion of HTR2A(Il5^(cre/+)·Htr2a^(flox/flox)mice)abolished the DOI-mediated inhibition of ILC2s in a papain-induced mouse model of *** conclusion,serotonin and DOI could restrict the type 2 lung immune response,indicating a potential treatment strategy for type 2 lung inflammation by targeting HTR2A on ST2+ILC2s.

关键词： Serotonin(5-HT) HTR2A DOl Group 2 innate lymphoid cell Type 2 lung inflammation

来源：评论

学校读者我要写书评

暂无评论

MSAT: Biologically Inspired Multi-Stage Adaptive Threshold for Conversion of Spiking Neural Networks

arXiv

引用

arXiv 2023年

作者： He, Xiang Li, Yang Zhao, Dongcheng Kong, Qingqun Zeng, Yi Beijing China School of Artificial Intelligence University of Chinese Academy of Sciences Beijing China Brain-inspired Cognitive Intelligence Lab Institute of Automation Chinese Academy of Sciences Beijing China Center for Excellence in Brain Science and Intelligence Technology Chinese Academy of Sciences Shanghai China

Spiking Neural Networks (SNNs) can do inference with low power consumption due to their spike sparsity. ANN-SNN conversion is an efficient way to achieve deep SNNs by converting well-trained Artificial Neural Networks (ANNs). However, the existing methods commonly use constant threshold for conversion, which prevents neurons from rapidly delivering spikes to deeper layers and causes high time delay. In addition, the same response for different inputs may result in information loss during the information transmission. Inspired by the biological model mechanism, we propose a multi-stage adaptive threshold (MSAT). Specifically, for each neuron, the dynamic threshold varies with firing history and input properties and is positively correlated with the average membrane potential and negatively correlated with the rate of depolarization. The self-adaptation to membrane potential and input allows a timely adjustment of the threshold to fire spike faster and transmit more information. Moreover, we analyze the Spikes of Inactivated Neurons error which is pervasive in early time steps and propose spike confidence accordingly as a measurement of confidence about the neurons that correctly deliver spikes. We use such spike confidence in early time steps to determine whether to elicit spike to alleviate this error. Combined with the proposed method, we examine the performance on non-trivial datasets CIFAR-10, CIFAR-100, and ImageNet. We also conduct sentiment classification and speech recognition experiments on the IDBM and Google speech commands datasets respectively. Experiments show near-lossless and lower latency ANN-SNN conversion. To the best of our knowledge, this is the first time to build a biologically inspired multi-stage adaptive threshold for converted SNN, with comparable performance to state-of-the-art methods while improving energy efficiency. Copyright © 2023, The Authors. All rights reserved.

关键词： Energy efficiency

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：