如传统的监督学习中，数据样本是一些带有标签的静止训练集，只要样本数据之

For example, in traditional supervised learning, the data samples are some labeled static training sets. As long as the difference between the sample data is sufficiently obvious, a good model can be trained. Reinforcement learning is a continuous decision-making process. In reinforcement learning, the agent does not have direct guidance information, but revises its own strategy through immediate feedback from the environment. The agent needs to continuously interact with the environment in real time. Get training data

As in traditional supervised learning, data samples are labeled static training sets that can train a good model as long as the differences between sample data are obvious enough. Intensive learning is a continuous decision-making process, in intensive learning, the intelligent body does not have direct guidance information, but through the immediate return of environmental feedback to correct their own strategies, intelligent body needs to constantly interact with the environment in order to obtain training data in real time

For example, in the traditional supervised learning, the data samples are static training sets with labels. As long as the differences between the sample data are obvious enough, a good model can be trained. Reinforcement learning is a continuous decision-making process. In reinforcement learning, agents do not have direct guidance information, but modify their own strategies through the immediate return of environmental feedback. Agents need to constantly interact with the environment to obtain training data in real time<br>