1) State 2) Action and then 3) Reward diagram: Which ML approach to use?

Question

It is looks like a reinforcement learning diagram however it's slightly different. I'll explain the numbers.

• 1) The environment first gives the agent a state
• 2) The agent does it's magic and then returns an action
• 3) The environment decides if this action is to be rewarded or punished for that action which depended on the state first sent

Is there a ML approach that satisfys this?

Answer 1

Can't you just formulate this as some loss over $\hat{y} = f(\mathbf{X})$ where $\mathbf{X}$ is the state (input vector), $\hat{y}$ is the action (prediction) and the loss is how you decide to reward/punish?

If you can indeed formulate it like this, you basically want to identify $f$, which sounds like standard regression (continuous $\hat{y}$) or classification (discrete $\hat{y}$) to me.