A Multiphase Semistatic Training Method for Swarm Confrontation Using Multiagent Deep Reinforcement Learning

<div>Individual and group reward. (a) Individual reward of single-phase training. (b) Individual reward of multi-phase training. (c) Group reward of single-phase training. (d) Group reward of multi-phase training.</div>

Computational Intelligence and Neuroscience

fig5

Figure 5

Figure 5: A Multiphase Semistatic Training Method for Swarm Confrontation Using Multiagent Deep Reinforcement Learning 