University of Cambridge > Talks.cam > Churchill CompSci Talks > Adversarial attacks against reinforcement learning agents

  Browse A-Z

Log in

Googledetails Microsoftdetails

🔐 Log In

Information on

Subscribing to talksdetails Finding a talkdetails Adding a talkdetails Disseminating talksdetails Help and Documentationdetails

ℹī¸ Information

Adversarial attacks against reinforcement learning agents

Download to your calendar using vCal

If you have a question about this talk, please contact Matthew Ireland .

Reinforcement learning agents have progressed by leaps and bounds over the last few years. We now have self-driving cars from Tesla, competitive game playing agents from OpenAI and Deepmind and many more.

While our knowledge about such technologies has improved, so too has our knowledge about the inherent weaknesses in such systems.

In this presentation, I will be discussing how reinforcement learning agents work, how adversarial attacks take advantage of weaknesses to fool these agents, and some of the defences we have against such attacks.

This talk is part of the Churchill CompSci Talks series.

This talk is included in these lists:

Note that ex-directory lists are not shown.

Other lists

SG1 Life Sciences & Society Average on weekdays: 9027.21326165 kWh

Other talks

Group Work Afternoon Break Peri-Operative Medicine, and Obstetrics and Gynaecology Cambridge RNA Club - ONLINE Christopher Cowie on the epistemology of assessing biosignature claims Am I a Member? Auditing Private Machine Learning
 

Š 2006-2025 Talks.cam, University of Cambridge. Contact Us | Help and Documentation | Privacy and Publicity