D61+Live 2018 – Building robust AI
While artificial intelligence can undertake some tasks more efficiently than humans, there are other tasks it cannot adequately replicate. What happens if an autonomous vehicle interprets a tampered stop sign as a speed limit? What if Alexa or Siri can be tricked into recording our conversations? In this session at D61+ Live Brisbane 2018, we spoke to AI experts about the concept of adversarial machine learning and how we can improve the robustness of these systems to prevent these things from happening.
On this panel:
Associate Professor of Computer Science
School of Electrical Engineering and Computer Science
Group Leader
Machine Learning, CSIRO’s Data61
Moderator: Ketan Joshi, Communications Advisor, CSIRO’s Data61
Previous post:
Coming up next: