You can equally define over states in action pairs.

So the planning problem is now the problem which relies on an action to each possible state.

The pi that we want the optimal policy is the one that's going to maximize the discounted, total Reward.

Everybody dreams.

I'm in it for the money.

There was no salary.

What do we do?

Bringing me back would've gotten you twice that.

I will give you the best criminal defense that money can buy.

And the offered them, for performance, three levels of rewards small reward, medium reward, large reward.

56 billion hours work.

(Laughter)

What about my account?

People offered the medium level of rewards did no better than people offered the small rewards.

You make three... we take one.

Your reward is your life.

Added incentive.

That's the saving grace of a covert operation.

This droid is worth more I get paid more.

And the price is 100.

Was it?

Is there, you know, remuneration?

It's certainly the people you work with.

Ain't nothing free in my world.

Added incentive.

He's being paid.

This is essentially, a typical motivation scheme within organizations, right?

This is more than I have earned.

We're sitting on double shares. 2o,ooo apiece!

Another fee!

Apart from his pay, which is considerable...

And what exactly do you want in return?

With a price that Scylla commands on the open market,

How much are they paying you, Jack?

One

So, what do i get out of this mystety deal, man?

The layoffs and payroll reductions.

I'll get a reward.

The money was good.

If... you ask me one more question...

Our laws then indulge this bias with ever broadening protections and massive rewards.

With 0.1 chance, we'll stay in the same state, in which case the value is 3.

What is your usual salary?

And I think the most interesting way to think about how all this is going on is in terms of rewards.

Well what happened though, was that, the people offered the medium reward, did no better than the people offered the small reward, but this time around, the people offered the top reward, they did worst of all.