Note 1: If you are looking for a solution to a problem. Do not start with googling it straight away. Understand the problem and come up with a solution of your own, if you can, or at least spend time brain storming the problem or the solution you came up with. For a day or two, work on something else. After this read the literature related and how others are solving it. Lastly, watch lectures/talks on the problem/topic and if you can, please talk to someone knowledgable in this area in person. Ask for his/her opinion and share yours. After this all you need to do is read...read....read....and......read.

Here is a list of links that I found intersting:

1. The bitter leasson--by Rich sutton
2. On the importance of the independent and identically distributed (i.i.d.) assumption in statistical learning
3. Blogs of Guilherme Kunigami
4. Notes on Noise Contrastive Estimation and Negative Sampling
5. A Gentle Introduction to Noise Contrastive Estimation
6. GMM
7. Understanding self-supervised contrastive learning