:::

News & Events

:::

Speaker:Dr. Pin-Yu Chen(Chief Scientist, RPI-IBM AI Research Center, IBM Research Research Staff Member, Trusted AI Group, IBM Research AI, USA)

Seminar
Poster:Webmasters事件日期:2020-01-15
  
Topic:Adversarial Robustness for Deep Learning: Trends and Challenges   
 
Speaker:Dr. Pin-Yu Chen(Chief Scientist, RPI-IBM AI Research Center, IBM Research Research Staff Member, Trusted AI Group, IBM Research AI, USA)
 
Date Time:WED. Jan 15,2020, 10:40 AM - 11:30 AM 
 
Place: 4F-427, Assembly Building I
 
Abstract

As deep neural networks have achieved unprecedented performance on a variety of datasets and across domains, developing proper and efficient robustness measures and better training algorithms that are capable of strengthening model robustness is the next crucial milestone toward trustworthy and reliable machine learning systems. This talk will consist of two parts. In the first part, we will provide an overview of attack-specific and attack-independent methods for evaluating adversarial robustness of neural networks. In the second part, we will introduce our recent works on detecting adversarial inputs and new methods that aim to improve adversarial robustness. Finally, we will conclude the talk with some ongoing challenges and potential applications using adversarial machine learning.
 
Last modification time:2020-01-15 AM 11:12

  • recruiting animation-EN
  • NCTU
cron web_use_log