How to learn machine learning and deep learning (AI) based on your high school maths

I am launching a new initiative called the Countdown Institute in August. 

Welcome thoughts and feedback 

We help you to understand machine learning and deep learning(AI) based on your high school maths concepts. This allows you to transition your career towards AI based on your existing knowledge.

Machine learning and Deep Learning are complex concepts but you can understand them based on your high school maths 

What do we mean by that?

Essentially, till the year 17/ 18, you learn four components i.e.

• Probability Theory
• Statistics
• Linear Algebra
• Optimization

These ideas are definitely known to you if you have studied a maths/ science based bachelors or engineering programs

These four ideas also underpin machine learning and deep learning 

The objective of our work is to build a bridge between these high school maths concepts and data science (which is a collective term for machine learning and deep learning) 

Why is this approach unique?

Instead of AI, let’s talk instead of Elon Musk!

Elon Musk talks of first principles thinking 

That means you boil down a process to it’s fundamental parts and work backwards through each building block to connect to the essential elements. This needs more in-depth thinking. 

In other words, first principles thinking is the act of boiling a process down to the fundamental parts that you know are true and building up from there. When applied in this context(learning AI), it means to connect complex ideas back to simple ones especially to ideas that you already know

You may have tried to learn Machine learning and Deep Learning previously with limited success. 

In my experience of teaching machine learning and deep learning, there are three problems for effective learning. 

Often, these are missed by practitioners but are very familiar to teachers - especially if they teach in class (face to face)

• Cognitive dependencies - ideas depend on other ideas which need to be explained first
• Cognitive overload - there is too much information. Most of it is good. But you don’t know which sequence to approach the topic. 
• First principles thinking - ie how things tie back to each other

Details of Topics and modules

Topics

Introduction to machine learning and deep learning 

Slaying the unicorn (narrowing the scope)

Ten themes roadmap

     1 functions and mathematical modelling

     2. Probability and inference 

     3. machine learning 

     4. deep learning

     5. model evaluation

     6. feature engineering 

     7. A taxonomy of algorithms

     8. Unsupervised learning

     9. Bayesian approaches

    10. A recap of ideas

Each topic has 10 to 15 modules each of 10/15 mins each so around 130 modules in total

Why now and what you get

• The total price is 99 USD + taxes as applicable 
• When you sign up, you get about a third of the modules - the next third in the next month and final third month after
• You get lifetime access to content and videos
• You can post questions online which I can answer - however, considering the accessible price point - there are limits to what I can respond 
•  The material is based on my teaching (but we are not affiliated to any educational institution)
• I intend to keep this group small and selective so we can learn from it 
• This is a fixed cost and it's very low to keep the material accessible globally. There are no extras and there is no refund
• Aim is to cover both the breadth and the depth 

Social enterprise

Finally, if you have known me over the years, I have always been interested in the social side of education. In this venture, we are working with the autism community to provide free access. 

Any questions, please contact me at ajit.jaokar at feynlabs.ai