In value iteration, we recursively calculate the value function so that in the end, we get what's called the optimal value function.

Whatever function value I get at the positive value of number,

I'm going to get the same function value at the negative value.

We want to return a particular value value corresponding to a function.

The P value is a direct, is directly a function of the T value.

Let's now implement something that calculates the value function.

Specifically, the value of the function test on a.

We have our debug function return a value named resume.

At t equals c, what's the value of the function?

Here is my solution, I have the value function initialized.

lets just think about the value of the function there.

So when t is equal to c, this value, the value of the function should be equivalent to the value of the original green function at zero, so it's equivalent to that value, which makes sense.

Something is only a function if for a given x value, you only have one y value.

In the square root function, the y variable always has the same value as the return value.

I print out the value function using these commands over here.

When we enter function 1, we print out current value of x in scope, the current value of y in scope, and then we call function 2.

For the state B3, calculate the value function assuming that we have a value function as shown over here and all the open states have a value of assumed 0, because we're still in the beginning of our value update.

If it's a function definition, then we're going to pull out the function name, the function parameters, the function body, and then we're going to make up a function value out of that.

At the end of absolute value or Fibonacci or factorial or almost any function or procedure we want to return with the final value, return an arbitrary expression that becomes the value of the function.

Often, we think of the function as return in just one value.

But leaving this concern aside, it actually computes the correct value function.

So you can actually get rid of that absolute value function there.

Well, it's going to be the value of the Dirac delta function.

The first part of our return value is just the word function.

So the minimum value that the function could take on is 1.

Our JavaScript interpreter is going to have to compute the value of 6 factorial, a function call within a function call within a function call.

Assuming that the value function is initialized with 0 everywhere, what will be the value after a single value iteration for the state up here.

It replaces the original input command function by a new function, which wraps around the original input command function, calls it, also returns its original value.

Here is my output for the value function. I can read those values.

It gets us the value function, and it gets us the policy action.

The height is the value of the function, which is x y squared.

So this value of greeter this function defined here will be copied back.

For example, if I modify the probability of success into a deterministic function, then my value function will look as follows.

Function 2 also prints out its value of x and its value of y, and down here we get the ball rolling by calling function 1 on Hong Kong in 1997.

Now, my cost function has got worse. cause it starts off from this value then now my value has gotten worse.