摘要:
1。 require的特点是仅加载一次,并且对于模块会按照特定的搜索规则去查找文件并加载。如果模块已经有了,不会重新再运行一次;没有的话,再加载一次。 dofile:读入代码文件并编译执行。每调用dofile一次,都会重新编译执行一次。 loadfile:编译代码,将整个模块文件当成一个函数返回,但 阅读全文
摘要:
Combat Task(对抗任务)CommNet We simulate a simple battle involving two opposing teams in a 15 * 5 grid as shown in Fig. 2(middle).Each team consists of m 阅读全文
摘要:
Human-level control through deep reinforcement learning PS:(结合论文Playing Atari with Deep Reinforcement Learning) 1. 论文背景 本文主要解决如何在高维度输入情况下进行增强学习Reinfor 阅读全文
摘要:
Learning Multiagent Communication with BackpropagationSainbayar Sukhbaatar, Arthur Szlam, Rob Fergus Many tasks in AI require the collaboration of mul 阅读全文