Workshop
Multi-Agent Security: Security as Key to AI Safety
Christian Schroeder de Witt · Hawra Milani · Klaudia Krawiecka · Swapneel Mehta · Carla Cremer · Martin Strohmeier
Room 223
Sat 16 Dec, 7 a.m. PST
This workshop proposal builds on the observation that the AI and cyber security communities are currently not sufficiently interconnected to navigate risks and opportunities in our multi-agent world. Through a series of discussions involving experts and audiences, provocation and intervention keynotes, and contributed content, we aim to compare, contrast, and synthesize near- and long-term perspectives of AI deployment across society. The fundamental goal of this workshop is to bring together researchers, practitioners, and activists across AI and cyber security in order to create a blueprint for the future of AI security in a multi-agent world, and to define, explore, and challenge the nascent field of multi-agent security (MASEC).
Submission deadline: September 25, 2023
Acceptance Notification: October 27, 2023
Workshop date: December 16, 2023
Schedule
Sat 7:00 a.m. - 7:10 a.m.
|
Opening Remarks
(
Moderated
)
>
SlidesLive Video |
🔗 |
Sat 7:10 a.m. - 7:35 a.m.
|
Multi-Agent Risks from Advanced AI
(
Keynote
)
>
SlidesLive Video |
Lewis Hammond 🔗 |
Sat 7:35 a.m. - 8:00 a.m.
|
Key Challenges in Foundation Models (... and some solutions!)
(
Keynote
)
>
SlidesLive Video |
Volkan Cevher 🔗 |
Sat 8:00 a.m. - 8:25 a.m.
|
Multi-Agent Vulnerabilities in Superhuman AI
(
Keynote Talk
)
>
SlidesLive Video |
Adam Gleave 🔗 |
Sat 8:24 a.m. - 8:25 a.m.
|
Towards AI-based auditing of privacy risks in privacy-enhancing technologies
(
[On-Demand] Keynote
)
>
SlidesLive Video |
Ana-Maria Cretu 🔗 |
Sat 8:24 a.m. - 8:25 a.m.
|
Recent Advances on Online Learning in Games.
(
[On-Demand] Keynote
)
>
SlidesLive Video |
Stratis Skoulakis 🔗 |
Sat 8:25 a.m. - 8:40 a.m.
|
Language Agents as Hackers: Evaluating Cybersecurity Skills with Capture the Flag
(
Oral
)
>
link
SlidesLive Video |
John Yang · Akshara Prabhakar · Shunyu Yao · Kexin Pei · Karthik Narasimhan 🔗 |
Sat 8:40 a.m. - 8:55 a.m.
|
Leading the Pack: N-player Opponent Shaping
(
Oral
)
>
link
SlidesLive Video |
Alexandra Souly · Timon Willi · Akbir Khan · Robert Kirk · Chris Lu · Edward Grefenstette · Tim Rocktäschel 🔗 |
Sat 9:05 a.m. - 9:20 a.m.
|
Cooperative AI via Decentralized Commitment Devices
(
Oral
)
>
link
SlidesLive Video |
Xyn Sun · Davide Crapis · Matt Stephenson · Jonathan Passerat-Palmbach 🔗 |
Sat 9:30 a.m. - 10:30 a.m.
|
Panel Debate: How can we make AI more secure?
(
Panel Debate
)
>
SlidesLive Video |
🔗 |
Sat 10:30 a.m. - 12:00 p.m.
|
Poster Session
(
Poster Session & Lunch break
)
>
|
🔗 |
Sat 12:10 p.m. - 12:25 p.m.
|
I See You! Robust Measurement of Adversarial Behavior
(
Oral
)
>
link
SlidesLive Video |
Lars Ankile · Matheus Xavier Ferreira · David Parkes 🔗 |
Sat 12:25 p.m. - 12:40 p.m.
|
Oracles & Followers: Stackelberg Equilibria in Deep Multi-Agent Reinforcement Learning
(
Oral
)
>
link
SlidesLive Video |
Matthias Gerstgrasser · David Parkes 🔗 |
Sat 12:40 p.m. - 12:50 p.m.
|
Defining and Mitigating Collusion in Multi-Agent Systems
(
Spotlight
)
>
link
SlidesLive Video |
Jack Foxabbott · Sam Deverett · Kaspar Senft · Samuel Dower · Lewis Hammond 🔗 |
Sat 12:50 p.m. - 1:00 p.m.
|
Second-order Jailbreaks: Generative Agents Successfully Manipulate Through an Intermediary
(
Spotlight
)
>
link
SlidesLive Video |
Mikhail Terekhov · Romain Graux · Eduardo Neville · Denis Rosset · Gabin Kolly 🔗 |
Sat 1:00 p.m. - 1:10 p.m.
|
Harnessing the Power of Federated Learning in Federated Contextual Bandits
(
Spotlight
)
>
link
SlidesLive Video |
Chengshuai Shi · Kun Yang · Ruida Zhou · Cong Shen 🔗 |
Sat 1:10 p.m. - 1:20 p.m.
|
Beyond Worst-case Attacks: Robust RL with Adaptive Defense via Non-dominated Policies ( Spotlight ) > link | Xiangyu Liu · Chenghao Deng · Yanchao Sun · Yongyuan Liang · Furong Huang 🔗 |
Sat 1:20 p.m. - 1:30 p.m.
|
Dynamics Model Based Adversarial Training For Competitive Reinforcement Learning
(
Spotlight
)
>
link
SlidesLive Video |
Xuan Chen · Guanhong Tao · Xiangyu Zhang 🔗 |
Sat 1:30 p.m. - 1:40 p.m.
|
RAVE: Enabling safety verification for realistic deep reinforcement learning systems
(
Spotlight
)
>
link
SlidesLive Video |
Wenbo Guo · Taesung Lee · Kevin Eykholt · Jiyong Jang 🔗 |
Sat 1:40 p.m. - 1:50 p.m.
|
Multiagent Simulators for Social Networks
(
Spotlight
)
>
link
SlidesLive Video |
Aditya Surve · Archit Rathod · Mokshit Surana · Gautam Malpani · Aneesh Shamraj · SAINATH SANKEPALLY · Raghav Jain · Swapneel Mehta 🔗 |
Sat 1:50 p.m. - 2:00 p.m.
|
Robustness to Multi-Modal Environment Uncertainty in MARL using Curriculum Learning
(
Spotlight
)
>
link
SlidesLive Video |
Aakriti Agrawal · Rohith Aralikatti · Yanchao Sun · Furong Huang 🔗 |
Sat 2:00 p.m. - 2:10 p.m.
|
Generation of Games for Opponent Model Differentiation
(
Spotlight
)
>
link
SlidesLive Video |
David Milec · Viliam Lisy · Christopher Kiekintveld 🔗 |
Sat 2:10 p.m. - 2:20 p.m.
|
Robust Q-Learning against State Perturbations: a Belief-Enriched Pessimistic Approach
(
Spotlight
)
>
link
SlidesLive Video |
Xiaolin Sun · Zizhan Zheng 🔗 |
Sat 2:20 p.m. - 2:30 p.m.
|
Stackelberg Games with Side Information
(
Spotlight
)
>
link
SlidesLive Video |
Keegan Harris · Steven Wu · Maria-Florina Balcan 🔗 |
Sat 2:30 p.m. - 2:40 p.m.
|
Assessing Risks of Using Autonomous Language Models in Military and Diplomatic Planning
(
Spotlight
)
>
link
SlidesLive Video |
Gabe Mukobi · Ann-Katrin Reuel · Juan-Pablo Rivera · Chandler Smith 🔗 |
Sat 2:40 p.m. - 2:50 p.m.
|
Decentralized agent-based modeling ( Spotlight ) > link | Ayush Chopra · Arnau Quera-Bofarull · Nurullah Giray Kuru · Ramesh Raskar 🔗 |
Sat 2:50 p.m. - 3:00 p.m.
|
Safe Equilibrium
(
Spotlight
)
>
link
SlidesLive Video |
Samuel Ganzfried 🔗 |
Sat 3:00 p.m. - 3:30 p.m.
|
Closing Remarks and Award Ceremony
(
Closing Remarks
)
>
SlidesLive Video |
🔗 |
-
|
Robustness to Multi-Modal Environment Uncertainty in MARL using Curriculum Learning ( Poster ) > link | Aakriti Agrawal · Rohith Aralikatti · Yanchao Sun · Furong Huang 🔗 |
-
|
Defining and Mitigating Collusion in Multi-Agent Systems ( Poster ) > link | Jack Foxabbott · Sam Deverett · Kaspar Senft · Samuel Dower · Lewis Hammond 🔗 |
-
|
Multiagent Simulators for Social Networks ( Poster ) > link | Aditya Surve · Archit Rathod · Mokshit Surana · Gautam Malpani · Aneesh Shamraj · SAINATH SANKEPALLY · Raghav Jain · Swapneel Mehta 🔗 |
-
|
Oracles & Followers: Stackelberg Equilibria in Deep Multi-Agent Reinforcement Learning ( Poster ) > link | Matthias Gerstgrasser · David Parkes 🔗 |
-
|
Dynamics Model Based Adversarial Training For Competitive Reinforcement Learning ( Poster ) > link | Xuan Chen · Guanhong Tao · Xiangyu Zhang 🔗 |
-
|
Beyond Worst-case Attacks: Robust RL with Adaptive Defense via Non-dominated Policies ( Poster ) > link | Xiangyu Liu · Chenghao Deng · Yanchao Sun · Yongyuan Liang · Furong Huang 🔗 |
-
|
Generation of Games for Opponent Model Differentiation ( Poster ) > link | David Milec · Viliam Lisy · Christopher Kiekintveld 🔗 |
-
|
Language Agents as Hackers: Evaluating Cybersecurity Skills with Capture the Flag ( Poster ) > link | John Yang · Akshara Prabhakar · Shunyu Yao · Kexin Pei · Karthik Narasimhan 🔗 |
-
|
Second-order Jailbreaks: Generative Agents Successfully Manipulate Through an Intermediary ( Poster ) > link | Mikhail Terekhov · Romain Graux · Eduardo Neville · Denis Rosset · Gabin Kolly 🔗 |
-
|
RAVE: Enabling safety verification for realistic deep reinforcement learning systems ( Poster ) > link | Wenbo Guo · Taesung Lee · Kevin Eykholt · Jiyong Jang 🔗 |
-
|
Cooperative AI via Decentralized Commitment Devices ( Poster ) > link | Xyn Sun · Davide Crapis · Matt Stephenson · Jonathan Passerat-Palmbach 🔗 |
-
|
Robust Q-Learning against State Perturbations: a Belief-Enriched Pessimistic Approach ( Poster ) > link | Xiaolin Sun · Zizhan Zheng 🔗 |
-
|
Assessing Risks of Using Autonomous Language Models in Military and Diplomatic Planning ( Poster ) > link | Gabe Mukobi · Ann-Katrin Reuel · Juan-Pablo Rivera · Chandler Smith 🔗 |
-
|
Stackelberg Games with Side Information ( Poster ) > link | Keegan Harris · Steven Wu · Maria-Florina Balcan 🔗 |
-
|
Safe Equilibrium ( Poster ) > link | Samuel Ganzfried 🔗 |
-
|
Leading the Pack: N-player Opponent Shaping ( Poster ) > link | Alexandra Souly · Timon Willi · Akbir Khan · Robert Kirk · Chris Lu · Edward Grefenstette · Tim Rocktäschel 🔗 |
-
|
Harnessing the Power of Federated Learning in Federated Contextual Bandits ( Poster ) > link | Chengshuai Shi · Kun Yang · Ruida Zhou · Cong Shen 🔗 |
-
|
Decentralized agent-based modeling ( Poster ) > link | Ayush Chopra · Arnau Quera-Bofarull · Nurullah Giray Kuru · Ramesh Raskar 🔗 |
-
|
I See You! Robust Measurement of Adversarial Behavior ( Poster ) > link | Lars Ankile · Matheus Xavier Ferreira · David Parkes 🔗 |