I was reading a paper on Addressing function approximation Error in Actor critic methods where they explain a benefit of their proposed method as follows:

I wish to build an intuition of this effect with either a proof or an example. Thank you
I was reading a paper on Addressing function approximation Error in Actor critic methods where they explain a benefit of their proposed method as follows:

I wish to build an intuition of this effect with either a proof or an example. Thank you
Copyright © 2021 JogjaFile Inc.
As an example, think of two distributions (RVs) with the same mean. If distribution 1 has a larger variance than distribution 2, the data will be more spread out. Distribution 1 will likely have a smaller minimum and a larger maximum than distribution 2 because its values tend to be further from the mean.