强化学习算法 —— 时变基线 —— 时变基线的计算方式为：取批次内所有轨迹，对每个时间步的回报值做平均

资料：

https://digitalassets.lib.berkeley.edu/techreports/ucb/text/EECS-2016-217.pdf

时变基线

The time-dependent baseline was computed by averaging the return at each timestep over
the trajectories in the batch.

豆包AI生成：