Skip to main content

Featured

How the New York Police Department is Using Artificial Intelligence ? Answers (latest)

Use of AI in NYPD ( AI news) AI has been changing things in many industries and departments recently, here is how it has had an impact on the New York Police Department. If you wanna know  more about AI  and be updated an AI trends you can follow me on- Pintrest Instagram Tumblr When a syringe wielding drill their tried sticking up a home spot near Yankee Stadium,police figured out quickly that it wasn't a one-off. A man had also used syringe a few weeks earlier while stealing a drill at another home depot 11 km south in Manhattan.  The match,though,wasn't made by an officer looking through files. It was done by pattern-recognition computer software developed by the New York Police Deparment. The software ,dubbed Paternizr, allow crime analysts stationed in each of the department's 77 precincts t compare robberies, larcenies and thefts t hundreds of thousands of crimes logged in the NYPD's database, transforming their hunt for crime pat
2 comments
Read more
Labels

Artificial Intelligence # What is the timing penalty for the AI?






Before I start I highly recommend you read my previous posts so as to have a better understanding.


.

In the previous tutorial we learnt the policy of AI or artificial Intelligence.
Now that the AI has a plan it can map out easily and reach the finish.Well, there is another layer coming in this post.Time factor!

Note-
Don't get confused between the value of the state and the penalty.
The value remains the same.

In order to make the AI finish the game or a given task in the right timing we add a penalty every time it changes state according to the environment and because of that the AI tries to finish the task or game with as less moves as possible.

Now let us understand the following wit examples of varied penalties.

Case 1- penalty of  0


This policy matches our original one.This can be applied only when you have surplus time and require the work to be done as safely as possible.

Case 2- penalty of -0.5
As you can see now the AI is taking a little risk so as to not accumulate - points by hitting the wall and be safe .This can be used in fields where a little risk is accepted.

Case 3- penalty of -2


Here the penalty is more than the final reward so the AI would even jump to the fire as it is only -1
rather than accumulating more -points.This must be applied only in areas where the timing is the most important thing.


I hope you understood the concept.
Any doubts - just leave a comment
UNTIL THEN STAY EXCITED ABOUT AI


Labels:

Comments

Post a Comment