Thousands of articles and videos exist to learn Machine Learning. Whether it’s technical lessons or theoretical videos.
The problem? Most of it is aimed at :
- students who want to watch passively what Artificial Intelligence can do
- future AI teachers who want to know the theory behind Machine Learning
Don’t get me wrong, it’s great for Machine Learning!
One gives exposure to the field. The other provides theoretical knowledge training for tomorrow’s teachers.
But let’s be honest:
You are NEVER going to use these skills in a work environment.
As Einstein said, “Theory is when you know everything and nothing works.”
He’s right.
If you want to have concrete results (and have real value in the working world) you need to develop practical and memorable skills.
This is what I call REPLICABLE PROTOCOLS.
For example:
- Analyze a dataset
- Create a Deep Learning model
- Modify the right hyperparameters
- Optimize the model
These protocols are practical, and that’s a plus.
That means they’ll really come in handy when you’re faced with a real-world problem.
Even better: they are memorable protocols.
Unlike all the abstract tutorials for learning the activation function, of loss, which allows the optimization of the layer neuron of… you will EASILY remember them.
This is very good news because when you communicate your results to non-experts, you won’t look like a geek who spends all day behind his computer. They will understand what you are saying.
That’s why I use these replicable protocols:
- They are practical so you can use them and communicate them easily
- They’re memorable so you’ll always have them in mind whatever your situation