~

NIPS2020/main.tex

@@ -168,9 +168,10 @@ systems. However, even using the multi-agent systems, previous works at root are
 training one brain for communicating among connected sensors (agents), as they combine
 the neural networks of all agents in the training process.
 
-Roughly, previous works force the agents evolving for generating communication
-protocol by setting a target that requires the cooperation of multiple
-agents. These works can be classified into two categories based on the
+Roughly, to evolve a symbolic language, previous works force the
+agents to finish a set target through cooperation, which requires communication
+among the agents.
+These works can be classified into two categories based on the
 environment settings: \emph{referential games} and \emph{multi-agent
   reinforcement learning system}.