Part 1 Foundations of modern transformer models

Since their introduction at the 31st Conference on Neural Information Processing Systems (NIPS 2017), transformers have become a popular topic of interest. The transformer is a type of machine learning model that, unlike its predecessors, can understand the context of words in a sentence by focusing on different parts of the sentence simultaneously. Exactly this fact makes transformers highly efficient variety of natural language processing (NLP) tasks, including text classification, summarization, and generation. They are being extensively used across many applications, underscoring the importance of understanding these models in full depth.