1 problem

0 votes0 replies0 views
Conjecture on two-agent centralized training for reinforcement learning algorithms
Let $V^{*[2]}(\cdot)$ denote the value function used in the two-agent centralized training phase. Suppose that the condition of the paper's main theorem is satisfied. Two-agent cen…
Conjectureopencommunication-systems multi-agent-systems reinforcement-learningarXiv posted 2 days ago

Conjecture on two-agent centralized training for reinforcement learning algorithms