Skip to main content
Artificial Intelligence

Return to Question

Bumped by Community user
edited tags
Link
nbro
nbro
  • 41.4k
  • 12
  • 114
  • 205
Bumped by Community user
Bumped by Community user
Bumped by Community user
Bumped by Community user
Bumped by Community user
added 22 characters in body; edited tags; edited title
Source Link
nbro
nbro
  • 41.4k
  • 12
  • 114
  • 205

Difference What is the difference between return and expected return in reinforcement learning?

Is my understanding correct?

At a time step t$t$, for a state S_{t}$S_{t}$, the return is defined as the discounted cumulative reward from that time step t$t$.

If an agent is following a policy (which in itself is a probability distribution of choosing a next state S_{t+1}$S_{t+1}$ from S_{t}$S_{t}$), the agent wants to find the value at S_{t}$S_{t}$ by calculating sort of "weighted average" of all the returns from S_{t}.$S_{t}.$ This is called the expected return.

Is my understanding correct?

Difference between return and expected return in reinforcement learning

Is my understanding correct?

At a time step t, for a state S_{t}, the return is defined as discounted cumulative reward from that time step t.

If an agent is following a policy (which in itself is a probability distribution of choosing a next state S_{t+1} from S_{t}), the agent wants to find the value at S_{t} by calculating sort of "weighted average" of all the returns from S_{t}. This is called expected return.

What is the difference between return and expected return?

At a time step $t$, for a state $S_{t}$, the return is defined as the discounted cumulative reward from that time step $t$.

If an agent is following a policy (which in itself is a probability distribution of choosing a next state $S_{t+1}$ from $S_{t}$), the agent wants to find the value at $S_{t}$ by calculating sort of "weighted average" of all the returns from $S_{t}.$ This is called the expected return.

Is my understanding correct?

Source Link
digi philos
digi philos
  • 51
  • 1

Difference between return and expected return in reinforcement learning

Is my understanding correct?

At a time step t, for a state S_{t}, the return is defined as discounted cumulative reward from that time step t.

If an agent is following a policy (which in itself is a probability distribution of choosing a next state S_{t+1} from S_{t}), the agent wants to find the value at S_{t} by calculating sort of "weighted average" of all the returns from S_{t}. This is called expected return.