Success in creating AI would be the biggest event in human history. Unfortunately, it might also be the last, unless we learn how to avoid the risks

Machine learning and artificial intelligence have had grown in popularity significantly over the past years. With this growth comes benefits and consequences. This article introduces the basic concepts of machine learning with intuitive visuals and little or no math/calculations. The goal is to give one a feel for how machine learning works. Understanding will play a key role in the future of machine learning as not having a good understanding of what is happening can have dire consequences.

At its core machine learning is an algorithm (or a set of instructions) for updating parameters. For example, If I had a function that looked like this.

The goal of this function is to guess the price of the house when given square footage and to do it as accurately as possible. It takes an 

 as parameters but we do not know what these parameters are. This is where machine learning comes to the rescue. Without machine learning, we would pick an 

 by trial and error. We could pick them and then look at how well these two parameters work on a set of data. I have coded up a demo below so you can play the role of machine learning and see how well you can make this model perform. The goal is to tweak and change 

 and try to make the error as small as possible (we will get into how this is calculated later). The data that is being used for this is a 

. Go ahead, play the role of machine learning, tweak 

 in the form below and see how small you can make the error.

Iframe

If your anything like me you may have played around with 

 and looked at how the error was changing while you changed 

. If it was going down one might keep changing the parameter and if it was going up we might change it in the other direction. What we just did was not machine learning but it was very close. Instead of you updating the parameters based on your intuition an algorithm updates the parameters based on mathematical principles.

Although important these mathematical principles often trip people up and can be a barrier to entry to learning machine learning. Let's go over some of them to gain further insight into how machine learning works.

The function that is learned in the machine learning process. In the example we just looked at it was the 

. This function has parameters that are learned. In our case, it was 

A cost function is a function that lets provide a measure for how will that machine learning model is performing on a set of examples. The cost function takes a model as input and outputs how well it is doing. If our model is doing well on a set of examples the cost will below. If it is doing poorly the cost is high. This is also sometimes called the Loss or Error function. In our example, the cost function was Cost(x, h) = (h(price) - RealPice)^2. Where 

I know this is a lot to take in but hopefully, you are still following.

The model learns by calculating the gradient of the cost function. In basic terms, the gradient is how each parameter (

) changes in relation to the cost. A big change will mean it affects the cost a lot and a little change will mean it affects the cost a little. Similarly to how in our example we change 

 until we say a decrease in cost. The gradient is the same idea but expressed in terms of math, namely calculus.

Now that we have some basic intuition we can see the learning process as a series of steps

There is a lot we have glossed over but hopefully, this provides some basic insight into how machine learning works and demystifies some of the magic.

From the very basics, learn what machine learning is in a way that is easy to understand, intuitive, and accessible for all backgrounds. Even if you are a beginner, learn about what is a Machine Learning model, cost function, and gain an intuitive understanding of how a model learns.

As someone who is in their fourth year of a double major in math and computer science I want to answer the question Does a math degree help one in the field of software engineering?

Does a Math Degree Help in Software Engineering?

What is Markdown?

Nextjs is a meta framework for React. Nextjs includes static site generation, sever site rendering (ssr), api routes, and many other features. We will explore all these ideas and more in this blog.

What is nextjs?

An accessible intro to Machine Learning

I am currently in my fourth year of completing a double major in math and computer science at the University of Prince Edward Island. As such, I am in a unique position to comment on my thoughts on whether or not a math degree will help one in software engineering. I have also been working in software development positions both, part-time and full-time, for about two years.

In a math degree, you will start by learning calculus, linear algebra, and other algorithmic-based learning. As you progress, most of the math becomes 

. Generally, you are given a set of axioms (true statements) and using those you have to contract some proof to show some proposition to be true. This forces you to think outside the box as no two problems are the same. You can't just apply an algorithm to solve your problem. Some of these courses are 

If you are going into university and wondering if a math degree will be useful for a future in software engineering, in terms of practical use, the answer is, probably not. If your goal is to be a general software engineer and write code for web dev or a similar field that you most likely will not use any of the facts or knowledge that a math degree offers. There are some exceptions. If you think you might want a future in machine learning, cryptography, research, programming software bundlers, and the list goes on and on. These are some of the many fields where you will use the knowledge and skills of a math degree in practice. If you are thinking of going into one of these fields, a math degree or possibly a minor will help you out tremendously. I have used many skills I have learned in my math courses as I complete my honours this year in machine learning.

Does it help you work as a software engineer?

Math courses challenge one in a way that traditional computer science courses do not. In math courses, specifically upper-year proof-based courses (which are almost all of the higher-level math courses), really challenges one's problem-solving skills in a way that you don't see in typical computer science courses. For example, as I talked about before making a proof comes down to when given a set of axioms (truth facts) you must manipulate those facts to prove a given proposition. This is a skill that is rarely used in your computer science courses. From my experience, a lot of CS courses are about memorizing facts or algorithms and then using those facts or algorithms through coding assignments, general assignments and exams. This type of learning rarely comes up in a software engineering position. No one is going to ask you to memorize the typescript syntax and then test you on it. You may learn it by using it but no one is going to sit you down and drill you with obscure abuse questions on the subject. What you will have to do a lot is problem-solving. You will be given many problems or tasks and you will have to break them down, understand them and finally solve them. This sounds simple but in practice can be very challenging. Math courses challenge you in a way that is similar to this. Many times, you are given a question that you have never seen anything like it before and are forced to practise and learn problem-solving techniques and skills.

If you are going into university and are faced with picking a math degree or computer science this can be a hard choice. A math degree will teach you problem-solving skills and some practical skills depending on what kind of field in software you wish to enter. A CS degree will probably teach you more practical skills for software engineering but may lack problem-solving skills and depth of knowledge for some fields. Overall, you should pick the field that interests you most. If you think you like math more, pick that, and you think programming and computers sound more interesting pick that. Overall, if you do what interests you more you will find the field or job that will be more enjoyable.

Thanks for hearing my thoughts, Logan Anderson

In this post, I will be going over how to make a 

 in tinaCMS. In my case, I want to make an action so that I could delete blog posts from my website.

A form action in TinaCMS is just a react component that loads within the alt menu of the form. See below.

 and a GitHub backend. This blog is hosted on vercel and takes advantage of there 

Note: for my examples, I will be using typescript. But everything should also work in javascript (just take out the types)

Let's say we wanted a picture of a lama to show up in our actions (because who doesn't like lamas). This would be pretty simple let's just create a react component that contains an image of a lama.

Note for other frameworks we may have to switch 

First, we are going to use the tinaCMS buttons and modal to the skelton of our delete action.

Now we have a popup and button, lets make it do something.

Next, we are going to get access to our tinaCMS client and call delete on it when the 

button is clicked. To do this we are going to use a factory function to allow us to change the way we access the form title and path.

Since we are using a factory function our blog file will look slightly different.

Look at that! we just made a pretty cool action and it was pretty simple to do so!

How to add a new action to a tinaCMS form. This brief tutorial will show you how to make a new action specifically,  a delete action.

Hi 👋 Im Logan!
I enjoy _

Latest Blog Posts

An accessible intro to Machine Learning

Does a Math Degree Help in Software Engineering?

How to add a new action to a tinaCMS form

Hi 👋 Im Logan! I enjoy _

Latest Blog Posts

An accessible intro to Machine Learning

Does a Math Degree Help in Software Engineering?

How to add a new action to a tinaCMS form

Hi 👋 Im Logan!
I enjoy _