100 Days of ML — Day 13 — Skills Needed to Learn AI at Every Level

As I remarked in the opening post of this journey, I taught myself AI, Machine Learning, and Deep Learning in 30 days. Having said that, I did have the math and programming background to dive as deep as I did, but I made a lot of mistakes that cut into my time. So just follow this path and you’ll be way more efficient.


How AI works

Step 1: Wrangle data

Step 2: Organize into rows and columns

Step 3: Data goes through weights then hidden layers then weights to outputs

Step 4: Backpropagation and gradient descent. Weights update back to hidden layers back to weights back to inputs.

Step 5: Repeat until error is less than 1%.



Pick a language: Python, R, C++, C, C#, Java, JavaScript. Then, learn how to use libraries. Now learn TensorFlow, Torch, or Keras. Cool.



You need to know probability and the error methods: Sum of Squares Errors, Mean Squared Errors.


You need to understand partial derivatives. To understand partial derivatives, you need to know derivatives.

This will aid in understanding of gradient descent and weight updates.

Linear Algebra

To understand Linear Algebra, you need to know what a tensor is.

To understand that, you need to know what a matrix is.

To understand that, you need to know what a vector is.

To understand that, you need to know what a scalar is.

So, a scalar is a single number: 5. We’ll throw it in brackets. [5].

A vector is a list of scalars: [1,2,3,4,5].

A matrix is a vector of vectors: [[1,2,3],[4,5,6],[7,8,9]].

A tensor is…

Go on.

Yep, a matrix of matrices and it’s better to see visually.

I need graphic design money

The other thing you need to know is matrix multiplication. It deserves an article of it’s own, but here’s the Math is Fun version:

Jimmy Murray is a Florida based comedian who studied Marketing and Film before finding himself homeless. Resourceful, he taught himself coding, which led to a ton of opportunities in many fields, the most recent of which is coding away his podcast editing. His entrepreneurial skills and love of automation have led to a sheer love of all things related to AI.





Source: Deep Learning on Medium