How to validate my knowledge of models implementation, pros and cons and area of applicability?

Question

So I've been doing ML for ~2 years in industry, I'm a BSc in applied math, finished several courses on ML/DL on coursera, read some specific topics in ML/DL books. Seem to be in the know, more or less.

But the catch is: I've never really validated my knowledge, as in taken an exam or certification. Some time has passed since I read theory and I sort of think that I'm losing it (especially since all I did was time-series forecasting).

What are the possible ways to validate that I actually understand stuff? Do I just text random ML experts "hey could I tell you stuff and you say whether that's bs or not"?..

Answer 1

I do not have a definite answer for your question, but one way to validate your ML knowledge is to attempt certain problems that are asked in ML interviews for jobs. It might be a good starting point although it will not cover all the relevant concepts. Two such resources are

• The questions section in Chip Huyen's ML Interviews book
• A Deep Learning Interviews book that contains many questions testing your ML and DL knowledge.

If you are looking to solve more theory questions, you can work through the exercises of Bishop's Pattern Recognition and Machine Learning book, although I am not sure if the solutions are available online.

Answer 2

1. Find a high quality and well recognized course, you can pay extra for a Certificate:

   • https://ocw.mit.edu/search/?t=Artificial+Intelligence

   • https://www.edx.org/course/the-analytics-edge

2. Try answering some of these questions, network with others:

   • https://mathoverflow.net/

3. Write a few papers, publish, and try to obtain an endorsement.