Oubo Ma
Papers
1
Total Citations
6
H-Index
1
About
Oubo Ma is a rising researcher in artificial intelligence, with a primary focus on the security and robustness of multi-agent reinforcement learning (MARL) systems. His work addresses critical vulnerabilities in partially observed environments, where agents must coordinate under limited information—a setting common to real-world applications like drone swarms and robotic manipulation. In his highly cited 2024 paper, “SUB-PLAY: Adversarial Policies against Partially Observed Multi-Agent Reinforcement Learning Systems,” Ma introduced novel adversarial attack strategies that expose how subtle policy manipulations can destabilize cooperative MARL agents, even when attackers have only partial observations. This contribution, already garnering 6 citations in a short time, highlights the urgent need for defensive mechanisms in deployed multi-agent systems. By bridging the gap between theoretical security threats and practical deployment risks, Ma’s work is shaping how the community approaches safe AI. His research not only advances adversarial machine learning but also provides a foundation for building more resilient autonomous systems, making him a notable voice in the emerging field of MARL security.
Research Focus
Key Achievements
Top Papers
- 1