论文标题
深入增强学习的抽象
Abstraction for Deep Reinforcement Learning
论文作者
论文摘要
我们表征了在深厚的增强学习背景下抽象的问题。在这个问题上可能会带来各种相似推理和关联记忆的良好方法，但是由于需要端到端的可不同性，它们带来了困难。我们审查了AI和机器学习的发展，可以促进其采用。
We characterise the problem of abstraction in the context of deep reinforcement learning. Various well established approaches to analogical reasoning and associative memory might be brought to bear on this issue, but they present difficulties because of the need for end-to-end differentiability. We review developments in AI and machine learning that could facilitate their adoption.
