Talk 1: 1900HRs - 1945HRs
Title: Defending Singapore’s Digital Borders
Description
Cyberspace will be increasingly central to maintaining Singapore’s economic success and geopolitical relevance. As one of the most connected countries globally, Singapore’s connectivity and digitalisation will only continue to intensify under our Smart Nation initiative. At the same time, cyber threats continue to proliferate. We see a variety of cyber threats from terrorists, criminal, and state-sponsored organisations, especially as Covid-19 has forced many of us to work, learn, and play from home, on our mobile and electronic devices.
MINDEF/SAF is responsible for enhancing Singapore’s security and sovereignty, no matter where the threats come from. Join us to find out more!
Bio
COL Justiin Ang, is Head, Cyber Strategy and Policy Department, of the Defence Cyber Organisation. COL Ang serves as the principal advisor to the Defence Cyber Chief for all matters pertaining to development and implementation of cyber strategies and plans. Prior to this, COL Justiin Ang held the appointment of Director (Strategic Policy, Policy-Operations and Futures) in the Defence Policy Office, where he oversaw the development of strategies to guide MINDEF’s defence relations, as well as policy matters pertaining to the SAF’s operations. COL Justiin Ang joined the Defence Cyber Organisation in 2020 after graduating from Carnegie Mellon University with a Masters in Science in Information Security. An Army infantry officer by vocation, he has experience in policy, operations, and talent management, having served with MINDEF/SAF for more than 20 years.
Talk 2: 1945HRs - 2030HRs
Title: Adversarial Attack on Machine Learning Models
Description
Machine learning models, especially deep neural networks, have achieved excellent performance in tasks such as image classification, object detection and natural language processing. Nonetheless, most machine learning models are vulnerable towards attacks using adversarial samples. A smallest perturbation on the input, such as change of a pixel’s value, could cause the model to give the wrong result. In this talk, I will give a brief introduction about some of the adversarial attacks and discuss what we can do to mitigate them.
Bio
Jiyi graduated from NUS School of Computing in 2018 with first class honors. Currently, he is a part-time PhD student in NUS, working with Prof. Chang Ee-Chien on topics in machine learning security. He is also working as a research scientist in PayPal.