www.premilab.com

HomePage

News

Members

Research

Projects

Publications

Patents

Seminar

Downloads

Opening

Album

UHAVE Workshop

Search »


International Workshop on Understanding and Harnessing AdVersarial Examples (U-HAVE)

RSS

                                         In conjunction with IEEE ICDM 2019, Beijing, 11/8-11/11

Adversarial examples refer to augmented data points generated by imperceptible perturbation of input samples. Being difficult to distinguish from real examples, such adversarial examples could however change the prediction of many of the best machine learning and data mining models including the state-of-the-art deep learning models. Fig.1 shows an illustration of adversarial examples. Adversarial examples have attracted much interest in machine learning and data mining recently. Various studies have been made on why the adversarial examples happen, how adversarial examples can be defended, and what can be done to harness adversarial examples for improving the robustness and accuracy of learning and mining algorithms.

                                                  
                                     center

Fig.1. Illustration of adversarial example. A panda image would be misclassified by state-of-the-art deep neural networks as a gibbon with a high confidence after adversarial noise is added (from Goodfellow 2014).

The workshop aims to provide professionals, researchers, and technologists with a single forum where they can exchange, discuss, and share the state-of-the-art theories and applications of adversarial examples particularly in deep neural networks and data mining approaches.

Program

1st International Workshop on Understanding and Harnessing AdVersarial Examples (UHAVE)
In conjunction with ICDM 2019, Room 401, November 8, 2019


13:30-14:20 Invited Talk: Adversarial Attacks in Brain-Computer Interfaces (BCIs), Professor Dongrui Wu, Huazhong University of Science and Technology
14:20-14:40 Deep Minimax Probability Machine, Lirong He, Ziyi Guo et al.
14:40-15:00 Application of Adversarial Examples in Communication Modulation Classification, Da Ke, Zhitao Huang, Xiang Wang et al.
15:00-15:20 Primitives Generation Policy Learning without Catastrophic Forgetting for Robotic Manipulation Fangzhou Xiong, Zhiyong Liu et al.

15:20-15:40 Coffee Break

15:40-16:20 Invited Talk (TBD)
16:20-16:40 Feature-Matching Speech Denoising GANs via Progressive Training , Li Zhao, Fang Zhou et al.
16:40-17:00 SimpleGAN: Stabilizing Generative Adversarial Networks with Simple Distributions, Shufei Zhang, Zhuang Qian et al.
17:00-17:20 Video Saliency Detection with Gated CNN and Residual Architecture, Hang Yu, Derong Chen, and Jiulu Gong
17:20-17:40 A two-stage multi-task learning-based method for selective unsupervised domain adaptation, Yong Dai, Jian Zhang, Shuai Yuan, and Zenglin Xu

Click here to download the detailed UHAVE19 Program

Duration: Half Day

====Topic of Interest====
Topics of interest include, but not limited to, the following aspects:

  • Foundations of understanding adversarial examples
    • Property analysis for adversarial examples
    • Explainable adversarial examples
    • Visualization of adversarial examples
    • Robust analysis of adversarial examples
    • Generalization analysis of adversarial examples
    • Connection of adversarial examples with regularization
    • Distributional robust optimization theory for adversarial examples

  • Theory and algorithms for generating adversarial examples
    • Generation of audio adversarial examples
    • Generation of visual adversarial examples in images and videos
    • Generation of adversarial examples in text
    • Generation of adversarial examples in structured data
    • Generation of adversarial examples based on gradient regularized methods
    • Generation of adversarial examples based on distribution
    • Stochastic generation of adversarial examples
    • Region based generation of adversarial examples
    • Generation of adversarial examples in manifold space

  • Theory and algorithms of defending adversarial examples
    • Robust generative adversarial networks with adversarial examples
    • Robust supervised neural networks with adversarial examples
    • Robust semi-supervised deep learning with adversarial examples
    • Robust statistical learning models with adversarial examples
    • Robust data mining approaches with adversarial examples
    • Data approximation, dimensional reduction, clustering with adversarial examples
    • Learning techniques for multimodal data with adversarial examples

  • Novel applications of adversarial examples in data mining
    • Business data security with adversarial examples
    • Decision making with adversarial examples
    • Data augmentation with adversarial examples
    • Counterfactual reasoning with adversarial examples
    • Medical / health informatics with adversarial examples
    • Text mining with adversarial examples
    • Biological data analysis with adversarial examples
    • Graph data analysis with adversarial examples
    • Time-series prediction with adversarial examples
    • Biometric recognition with adversarial examples

  • Other related adversarial learning and mining methods
    • Adversarial learning applications and methods
    • GAN
    • Any other related adversarial methods


Submission and Key Dates:

Please submit your papers here by choosing the UHAVE workshop here at http://www.icdm2019.bigke.org.
  • Paper Submission: August 20, 2019 (August 7)
  • Paper Notification: September 4, 2019
  • Camera-ready deadline and copyright forms: September 8, 2019


Publication:

All the accepted papers will be published in ICDM workshop proceedings which will be indexed by EI. Selected papers will be invited to be published in ISI index journals

Organizers:

Kaizhu Huang, Professor, Xi’an Jiaotong-Liverpool University,
Bio: Kaizhu Huang is currently a Professor and Head, Department of Electrical and Electronic Engineering, Xi’an Jiaotong-Liverpool University, China. He is also the founding director of Suzhou Municipal Key Laboratory of Cognitive Computation and Applied Technology. Prof. Huang has been working in machine learning, neural information processing, and pattern recognition. He was the recipient of 2011 Asia Pacific Neural Network Society (APNNS) Younger Researcher Award. He also received Best Book Award in National Three 100 Competition 2009. He has published 8 books in Springer and over 140 international research papers including about 60 SCI-indexed international journals, e.g., in journals (JMLR, Neural Computation, IEEE T-PAMI, IEEE T-NNLS, IEEE T-BME, IEEE T-Cybernetics) and conferences (NIPS, IJCAI, SIGIR, UAI, CIKM, ICDM, ICML, ECML, CVPR). He serves as associated editors in three international journals and board member in three international book series. He has been sitting in the grant evaluation panels in Hong Kong RGC, Singapore AI programs, and NSFC, China. He served as chairs in many international conferences and workshops such as ICONIP, AAAI, ACML, ICDAR, ACPR, SDA, and DMC. His personal website can be seen in http://www.premilab.com/KaizhuHUANG.ashx.

Ping Guo, Professor, Beijing Normal University,
Bio: IEEE senior member, CCF senior member, Chair of IEEE CIS Beijing Chapter (2015-2016). His research interests include computational intelligence theory and its applications in pattern recognition, image processing, software reliability engineering, and astronomical data processing. He hold 6 patents and has published more than 300 papers, and two books: “Computational intelligence in software reliability engineering”, and “Image semantic analysis.” He received 2012 Beijing municipal government award of science and technology (third rank) entitled "regularization method and its application". Professor Guo received his master's degree in optics from the Department of physics, Peking University, and received his Ph.D degree from the Department of computer science and engineering, Chinese University Hong Kong. His personal home page can be seen in http://sss.bnu.edu.cn/~pguo.

Zenglin Xu, Professor, University of Electronic Science and Technology of China,
Bio: Zenglin Xu is a Professor in School of Computer Science and Engineering at University of Electronic Science and Technology of China(UESTC). He obtained his PhD in Computer Science and Engineering from the Chinese University of Hong Kong, and after that he worked at Max-Planck Institute for Informatics, Germany and Purdue University, USA. He is the founding director of the Statistical Machine Intelligence and LEarning (SMILE) Lab. His research interests include machine learning and its applications on social network analysis, health informatics, and cyber security analytics. He has published over 70 papers in prestigious journals and conferences such as NIPS, ICML, IJCAI, AAAI, IEEE PAMI, IEEE TNN, etc. He is a recipient of China Thousand Talents(Youth) Program. He is also the recipient of the APNNS young researcher award in 2016, and the best student paper honorable mention of AAAI 2015 and ACML 2016. Dr. Xu has been a PC member or reviewer to a number of top conferences such as NIPS, ICML, AAAI, IJCAI, etc. He currently serves as an associated editor to Neural Networks and Neurocomputing. His personal home page can be seen in http://smilelab.uestc.edu.cn/.

Yuan He, Alibaba Staff Algorithm Engineer, Alibaba Group.
Bio: Dr. Yuan He is a Staff Algorithm Engineer in the Security Department of Alibaba, and working on artificial intelligence for business security. His research interests include data mining, computer vision, pattern recognition and machine learning. Before joining Alibaba, He was a research manager at Fujitsu working on document analysis system. He received his B.S. and Ph.D. degrees from Tsinghua University.

Tentative program committee

  • Jianke Zhu, Zhejiang University, China
  • Yannis Goulermas, University of Liverpool, UK
  • Jinchang Ren, Strathclyde University, UK
  • Amir Hussain, Edinburgh Napier University, UK
  • Qiufeng Wang, Xi’an Jiaotong-Liverpool University, China
  • Rui Zhang, Xi’an Jiaotong-Liverpool University, China
  • Yanming Zhang, Institute of Automation, Chinese Academy of Sciences, China
  • Irwin King, The Chinese University of Hong Kong, China
  • Jinfeng Yi, DiDi AI Research Institute, China
  • Zhanxing Zhu, Peking University, China
  • Jiayu Zhou, Michigan State University, USA
  • Kun Zhang, Carnegie Mellon University, USA


Main Contact

Kaizhu Huang, Professor, Xi’an Jiaotong-Liverpool University
Email:
Mail Address: EE510A, Engineering Building, Xi’an Jiaotong-Liverpool University
Ren’ai Road, No. 111, SIP, Suzhou, 215123, Jiangsu Province, China
  名称 大小
- 1.png 135.38 KB
- UHAVE19_Program.pdf 487.85 KB
苏ICP备14059053号
Admin - 登录 - Edit