Here we will explain what probability is and why it is an important part of robotics especially in the areas of AI and machine learning.

Again there are a handful of mathematical concepts that are said to be essential if you are doing advanced ai work or if you are an engineer. These areas are calculus, linear algebra, statistics and probabilty.

This is because AI involves buckets and buckets of data….millions and millions of pieces of data that has to be organized and studied and made to make sense to make better decisions.

Your facial recognition software will look at thousands of faces an hour and compare one to the other and this is where probability will come in.

At some point, the facial recognition software which may be law enforcement software may be looking to find a murderer in a crowd.

It will narrow down the number of faces to look at to a handful eventually and it must use probability and other mathematical tools to deternmine the chance or probability that the face they are about to arrest is the correct person and that a very important mistake hasn’t been made.

So lets start by talking about probability in our everyday lives before we talk about the math behind it.

If you have ever flipped a coin or thought about winning a lottery or getting hit by lighting then you have been dealing in the area of probabilities.

Its all about trying or taking a stab at something.

When you flip a coin the chances or probability of getting a head is said to be one out of two or even steven as they say.

When you flip the coin once you have made a try. And since there are only two chances here either heads or tails then it is said that you have even odds of getting either one.

Now you might try and flip the coin 15 times and you just might get 15 heads in a row which is just real lucky but when you talk abot probability you have to do a lot of tries because in the end, it all levels out and you will be able to say with certainty that the odds or probability of getting a head after millions of tries is always one out of two or again, even steven.

Some might say that you will never win a lottery because you have just about the same chance or probability of being hit by lighting.

That means that you probably have a low chance like maybe one chance out of 50 million tries.

In probability theory we use simple numbers between 0 and 1 and use percentabes to describe the probablility of something happening or not happening.

You have 1 or 100% chance of something happening but you can not 1.5 or 150% chance of something happening.

You can have a 0 amount, or a 0 percent chance of something happening but you can’t have less than 0 percent chance of something happening.

Just like that old expression from the 50’s where a person would suggest that you couldn’t do something they might say that you have less than a chance of having a snowball in hell that you can do it !

An articvle called Doctor, AI in the August 1, 2021 edition o the Winnipeg Free Press newspaper quoted an associate professor of medical science at Brown University, Hamish Fraser about AI and probability.

He mentioned that the older models just a few years ago used simpler types of mathematics and ways of dealing with data with decision trees and decision tables but the more modern massive data models must deal with true machine learning techniques using complex probability models to predict medical outcomes.

The article was largely dealing with the use of AI based medical chatbots to take the load off medical screening staff during the first part o the Covid situation.

Now lets define the term statistics and talk about how it relates to robotics and in particular ai or data science.

Stats is said to be the science of learning from our data.

Statistics or stats is usually a first year course at university for both sceicne and business students since some of the theory can apply to both the scientific and business community.

According to our old friend Wikepdia stats is a branch of mathematics that is involved in the collection, organization, analysis, interpretation and presentation of data.

Now this sounds alot like the sub category of artificial intelligence simply called data science.

Remember we have AI and some subsets of that are machine learning and then deep learning and then data science.

Data science deals with massive amounts of data and people use advanced math to process this data and make intelligent decisions based on the results.