澳门葡亰娱乐场

学术讲座

首页 > 通知动态  > 学术讲座
2019年12月18日学术报告二则(Associate Prof Jun Shen, University of Wollongong;陈亮 副教授,中山大学)
2019年12月13日14时 人评论

报告题目1Computational Intelligence Applications in Challenging Disciplines

报告时间20191218(周三)下午14:30

报告地点澳门葡亰娱乐场B404会议室

报告人Jun Shen

报告人单位University of Wollongong

报告人简介

Associate Prof Jun Shen is currently postgraduate coordinator in School of Computing and IT at University of Wollongong after serving Research Committer Chair and Head of Postgraduate Studied in School of Information System and Technology. He is associate editor of four journals including SCI indexed Journal of Computational Intelligent Systems and Journal of Complex and Intelligent Systems. He is member of editorial board of other eight journals. He serves as PC chair for seven international conferences and PC member for more than 250 times. He is regular reviewer for more than 20 journals including many IEEE Transactions in computer sciences. Dr Shen has published more than 70 journal papers and 120 conference articles or book chapters, with supervision of 12 PhD students to completion. He has been supported by 25 research grants including ARC DP, NSFC. Prof Shen’s expertise includes big data and its applications in multiple disciplines such as education, transport, management and manufacturing. He has been IEEE NSW Section’s Education Chapter’s chair for more than 10 years and had served as panel member for AIS/ACM’s curriculum review of MSIS2016 representing Asia-Pacific universities. He was a visiting scholar at MIT, GaTech and UCI in the past 10 years. He is senior member of ACM and IEEE.

报告摘要

With the boom of AI and machine learning technologies, huge development of new algorithms and methods in areas such as cybersecurity and computer vision has seen human’s life much easier. However, there are some disciplines, such as education, medicine and transport, still lacking appropriate approaches to tackling challenging problems. This talk will introduce some research work carried out at Centre for Digital Transformation in University of Wollongong, where evolutionary algorithms, classical machine learning and deep learning methods are helping with some novel solutions. The underlying problems include cold start recommendation in micro learning, pathogen-pathogen protein interaction predictions and primary delay analysis in train networks.


报告题目2AI安全——面向图数据的对抗攻击与防御

报告时间20191218(周三)下午15:30

报告地点澳门葡亰娱乐场B404会议室

报告人陈亮

报告人单位中山大学

报告人简介

陈亮博士,中山大学数据科学与澳门葡亰娱乐场副教授,区块链与智能金融研究中心副主任,入选广东省“珠江人才计划”引进创新创业团队,担任国际期刊《International Journal of Web Science》副主编,SSYSF青年科学家论坛副主席,主持国家自然科学基金青年基金、国家自然科学基金区域联合基金重点项目(广东部分),参与国家重点研发计划项目若干。主要研究方向为推荐系统、图计算、对抗学习、AI安全和服务计算。在国内外重要刊物和会议上发表论文80余篇,近五年论文引用1000余次。

报告摘要

深度神经网络强大的表示学习能力使得其近年来在诸多领域取得了巨大的成功。然而,在其卓越性能的背后,深度神经网络作为一个黑箱模型,缺乏可解释性与鲁棒性,使得它易受到对抗攻击。目前已有的对抗攻防工作主要围绕图像、文本和语音等领域,缺乏对图数据的对抗攻防相关研究。报告将分析面向图数据的对抗攻防的重要性、挑战、已有工作,及我们在这方面的思考和一些进展。

邀请人:马于涛副教授、王健讲师


版权所有 ©澳门葡亰娱乐场官网-澳门新葡亰网站app下载 | copyright © 2008-2019 Macau Casino. All Rights Reserved.

Baidu
sogou