To successfully navigate its environment, an agent must construct and maintain representations of the other agents that it encounters. Such representations are useful for many tasks, but they are not without cost. As a result, agents must make decisions regarding how much information t
在Flappy Bird和Sonic The Hedgehog两个视觉环境中，我们对自监督方法进行了小规模研究。我们定量评估了从这些任务中学到的表示，以及它们对新的情况（如新的关卡和纹理）的可推广性。最后，我们通过可视化它们关注的环境部分评估了这些自监督特征。我们的结果表明了表示的效用高度依赖于环境的视觉和动态特征。