VALSE

VALSE 首页 活动通知 查看内容

20190717-17期 脑启发视觉模型 Brain Inspired Visual Model

2019-7-15 22:10| 发布者: 程一-计算所| 查看: 3893| 评论: 0

摘要: 报告时间:2019年7月17日(星期三)晚上20:00(北京时间)主题:脑启发视觉模型 Brain Inspired Visual Model报告主持人:左旺孟(哈尔滨工业大学)报告嘉宾:斯白露(北京师范大学)报告题目:Brain-inspiredIntell ...

报告时间:2019年7月17日(星期三)晚上20:00(北京时间)

主题:脑启发视觉模型 Brain Inspired Visual Model

报告主持人: 左旺孟(哈尔滨工业大学)


报告嘉宾:斯白露(北京师范大学

报告题目:Brain-inspiredIntelligence: A Paradigm for Next Generation AI


报告嘉宾:刘明霞(University of North Carolina at Chapel Hill)

报告题目:Neuroimage Analysis for Automated Brain Disease Diagnosis


Panel议题:

1. 脑与视觉学习的融合发展的可行路径与前景。

2. 计算机视觉模型有没有可能对脑机制探索有帮助?

3. 人脑理解与机器学习应该以什么方式互相推动?Learning for understanding brain还是Brain Inspired Visual Learning,还有没有其他的方式?

4. 脑科学研究可能会在生理、病理、心理和认知等不同层面上开展,这些层面是否都会以一定方式推动脑启发的视觉学习模型的发展?

5. 是否存在一些特定的视觉问题,更适合从脑启发的视觉学习的角度开展研究?


Panel嘉宾:

斯白露(北京师范大学)、刘明霞(University of North Carolina at Chapel Hill)、高跃(清华大学)、胡晓林(清华大学)、张道强(南京航空航天大学)


*欢迎大家在下方留言提出主题相关问题,主持人和panel嘉宾会从中选择若干热度高的问题加入panel议题!

报告嘉宾:斯白露(北京师范大学

报告时间:2019年7月17日(星期三)晚上20:00(北京时间)

报告题目:Brain-inspiredIntelligence: A Paradigm for Next Generation AI


报告人简介:

BailuSi is a professor at the School of Systems Science of Beijing NormalUniversity. He obtained the PhD degree in Theoretical Neurophysics from BremenUniversity, Germany in 2007. During 2008-2013 he was postdoctoral researchersin the Sector of Cognitive Neuroscience of the International School for Advanced Studies and the Department of Neurobiology of the Weizmann Instituteof Science, working on the computational mechanisms of the neural circuits of spatial memory. Before he joined BNU in 2018, he was a Principle Investigatorof the State Key Laboratory of Robotics in Shenyang Institute of Automation,Chinese Academy of Sciences. His research interest includes neural signalprocessing, brain-inspired computation and neurorobotics. He is a member in the Committee of Computational Neuroscience and Neuroengineering of the ChineseSociety of Neuroscience, the committee of Intelligence Interaction of the Chinese Association for Artificial Intelligence, the committee of Automation for Environment Perception and Protection of the Chinese Association of Automation.


个人主页:

http://www.brainair.cn


报告摘要:

Tounderstand intelligence is one of the ultimate questions for human beings.Traditional AI research is faced with challenges such as robustness,scalability and interpretability. Brain, as the only general intelligencesystem in nature, constitutes a blueprint for AI research to understand andcreate intelligence. In this talk, I will review recent progress in neuroscienceresearch, and discuss neural network models of perception and memory. Byunderstanding the dynamics and computational mechanisms of the neural circuits,it is possible to find a path to understand and create intelligence.


参考文献:

[1] Dongye Zhao, Bailu Si, and Fengzhen Tang. Unsupervised feature learning for visual place recognition in changing environments. In Proceedings of the 2019 International Joint Conference on Neural Networks. 2019.

[2] Taiping Zeng and Bailu Si. Cognitive mapping based on conjunctive representations of space and movement. Frontiers in Neurorobotics, 11:61, 2017.

报告嘉宾:刘明霞(University of North Carolina at Chapel Hill)

报告时间:2019年7月17日(星期三)晚上20:30(北京时间)

报告题目:Neuroimage Analysis for Automated Brain Disease Diagnosis


报告人简介:

Mingxia Liu is a Research Instructor of University of North Carolina at Chapel Hill. Her research focuses on machine learning and pattern recognition, with applications of Artificial Intelligence to studying aging and brain disorders. Her classification method was ranked #1 in the ISBI Challenge of “Classification of Normal versus Malignant Cells in B-ALL White Blood Cancer Microscopic Images” in 2019. She is the recipient of Outstanding Doctoral Dissertation Nomination Award from the Chinese Association for Artificial Intelligence, Outstanding Doctoral Dissertation Award from the Computer Society of Jiangsu Province, China, Travel Award of MICCAI 2016, and Travel Award of IAPR 2012. She has served as the Area Chair of MICCAI 2019, Co-Chair of MLMI 2018-2019 and Co-Chair of GLMI 2019. She is a Guest Editor of Journal of Neuroscience Methods Special Issue on Deep Learning Methods and Applications in Neuroimaging, Multimedia Tools and Applications Special Issue on Multimodal Data Fusion, Learning, and Application, and Neurocomputing Special Issue on Multimodal Media Data Understanding and Analytics. She is currently an Academic Editor for PLOS ONE.


个人主页:

http://mingxia.web.unc.edu/


报告摘要:

Multi-modal neuroimages facilitate the automated diagnosis of brain disorders by providing fundamental insights into neurodegenerative patterns of the human brain. Nevertheless, there are still lots of challenges need to be addressed. This talk will first present some of our recent work on brain disease analysis using structural and functional magnetic resonance (MR) images. We will also discuss how the heterogeneous and incomplete multi-modal neuroimaging data may benefit the automated diagnosis of brain diseases.


参考文献:

[1] Chunfeng Lian, Mingxia Liu, Jun Zhang, and Dinggang Shen, "Hierarchical Fully Convolutional Network for Joint Atrophy Localization and Alzheimer’s Disease Diagnosis using Structural MRI," IEEE Transactions on Pattern Analysis and Machine Intelligence, DOI: 10.1109/TPAMI.2018.2889096, 2019. 

[2] Yongsheng Pan, Mingxia Liu*, Chunfeng Lian, Yong Xia, and Dinggang Shen. "Disease-Image Specific Generative Adversarial Network for Brain Disease Diagnosis with Incomplete Multi-Modal Neuroimages," In the 22nd International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI 2019), Shenzhen, China, Oct. 13-17, 2019. 

Panel嘉宾:高跃(清华大学)


嘉宾简介:

清华大学长聘副教授、博士生导师。入选国家千人计划青年项目。近年来作为项目负责人承担国家重点研发计划、自然科学基金联合基金重点项目等多项课题,主要研究领域为立体视觉及脑科学,在Human Brain Mapping、MICCAI、IEEE Trans. on Medical Imaging、CVPR等国际期刊及会议发表论文100余篇,论文引用5000余次,由Elsevier出版英文专著两部。担任IEEE Trans. on Signal and Information Processing over Network及Journal of Visual Communication and Image Representation等国际期刊编委,担任MICCAI等国际会议领域主席。


个人主页:

http://www.gaoyue.org

Panel嘉宾:胡晓林(清华大学)


嘉宾简介:

清华大学计算机系副教授。2007年在香港中文大学获得自动化与辅助工程专业博士学位,然后在清华大学计算机系从事博士后研究,2009年留校任教至今。他的研究领域包括人工神经网络和计算神经科学,主要兴趣包括开发受脑启发的计算模型和揭示大脑处理视听觉信息的机制,在知名国际期刊和国际会议上发表论文70余篇。他是IEEE Transactions on Neural Networks and Learning Systems和Cognitive Neurodynamics的编委。曾带领学生在各种模式识别国际竞赛中获得过6次冠军和3次亚军。


个人主页:

www.xlhu.cn

Panel嘉宾:张道强(南京航空航天大学)


嘉宾简介:

南京航空航天大学计算机科学与技术学院/人工智能学院教授、博士生导师,副院长。入选国家优秀青年科学基金、“万人计划”青年拔尖人才项目。近年来作为课题负责人承担国家重点研发计划、国家自然科学基金、牛顿高级学者基金等多项课题,主要研究领域为机器学习、脑影像智能分析与脑疾病早期诊断,在TPAMI、TMI、TIP、Neuroimage、Human Brain Mapping、NIPS、MICCAI、KDD等国际期刊及会议发表论文100余篇,论文引用9000余次。研究成果获教育部自然科学二等奖1项(第一完成人)。担任Journal of The Franklin Institute、《自动化学报》等期刊编委,任中国图象图形学会理事、中国图学学会图学大数据专委会副主任、中国人工智能学会机器学习专委会常委、江苏省人工智能学会医学图像处理专委会主任等职务。

主持人:左旺孟(哈尔滨工业大学)


主持人简介:

左旺孟,哈尔滨工业大学计算机学院教授、博士生导师。主要从事图像增强与复原、图像编辑与生成、物体检测与目标跟踪、图像与视频分类等方面的研究。在CVPR/ICCV/ECCV等顶级会议和T-PAMI、IJCV及IEEE Trans.等期刊上发表论文90余篇。


个人主页:

http://homepage.hit.edu.cn/wangmengzuo


19-17期VALSE在线学术报告参与方式:


长按或扫描下方二维码,关注“VALSE”微信公众号(valse_wechat),后台回复“17期”,获取直播地址。



特别鸣谢本次Webinar主要组织者:

主办AC:左旺孟(哈尔滨工业大学)

协办AC:张兆翔(中科院自动化所)

责任AC:左旺孟(哈尔滨工业大学)


VALSE Webinar改版说明:

自2019年1月起,VALSE Webinar改革活动形式,由过去每次一个讲者的方式改为两种可能的形式:

1)Webinar专题研讨:每次活动有一个研讨主题,先邀请两位主题相关的优秀讲者做专题报告(每人30分钟),随后邀请额外的2~3位嘉宾共同就研讨主题进行讨论(30分钟)。

2)Webinar特邀报告:每次活动邀请一位资深专家主讲,就其在自己熟悉领域的科研工作进行系统深入的介绍,报告时间50分钟,主持人与主讲人互动10分钟,自由问答10分钟。


活动参与方式:

1、VALSE Webinar活动依托在线直播平台进行,活动时讲者会上传PPT或共享屏幕,听众可以看到Slides,听到讲者的语音,并通过聊天功能与讲者交互;

2、为参加活动,请关注VALSE微信公众号:valse_wechat 或加入VALSE QQ群(目前A、B、C、D、E、F、G、H、I群已满,除讲者等嘉宾外,只能申请加入VALSE J群,群号:734872379);

*注:申请加入VALSE QQ群时需验证姓名、单位和身份,缺一不可。入群后,请实名,姓名身份单位。身份:学校及科研单位人员T;企业研发I;博士D;硕士M。

3、在活动开始前5分钟左右,讲者会开启直播,听众点击直播链接即可参加活动,支持安装Windows系统的电脑、MAC电脑、手机等设备;

4、活动过程中,请不要说无关话语,以免影响活动正常进行;

5、活动过程中,如出现听不到或看不到视频等问题,建议退出再重新进入,一般都能解决问题;

6、建议务必在速度较快的网络上参加活动,优先采用有线网络连接;

7、VALSE微信公众号会在每周四发布下一周Webinar报告的通知及直播链接。

8、Webinar报告的PPT(经讲者允许后),会在VALSE官网每期报告通知的最下方更新[slides]

9、Webinar报告的视频(经讲者允许后),会更新在VALSE爱奇艺空间,请在爱奇艺关注Valse Webinar进行观看。


斯白露[slides]

刘明霞[slides]

小黑屋|手机版|Archiver|Vision And Learning SEminar

GMT+8, 2024-4-27 10:26 , Processed in 0.014481 second(s), 14 queries .

Powered by Discuz! X3.4

Copyright © 2001-2020, Tencent Cloud.

返回顶部