2 papers
cs.MA2026
Orchestrating Level- Policies Against Unknown Opponents in Partially-Observable Dynamic Games
Addison Kalanther, Sanika Bharvirkar, Daniel Bostwick +2
Level- reasoning generates a hierarchy of policies specialized to opponents with different reasoning levels. When an opponent's level is unknown, a common deployment rule estima…
cs.LG2026
NePPO: Near-Potential Policy Optimization for General-Sum Multi-Agent Reinforcement Learning
Addison Kalanther, Sanika Bharvirkar, Shankar Sastry +1
Multi-agent reinforcement learning (MARL) is increasingly used to design learning-enabled agents that interact in shared environments. However, training MARL algorithms in general-…