SDS 513: Transformers for Natural Language Processing — with Denis Rothman

Writing Method

Denis Rothman shares his unique approach to writing books, emphasizing that his motivation is not financial gain but rather the desire to share knowledge and connect with people globally. He explains that his books are often pre-written in his mind, allowing him to write rapidly once he begins. This method stems from years of contemplation and experience in AI, which he organizes into coherent narratives.

The book is written in my mind. Like, I'm watching TV, you know, and the book is just up there, and it's like a woman carrying a baby, and all of a sudden it's just a pain. I have to get it out of my system.

--- Denis Rothman

Rothman also highlights the importance of engaging with his audience, particularly on platforms like LinkedIn, where he answers questions and shares insights from his extensive career 1 2.

   

AI Books

Denis Rothman's books cover a wide range of AI topics, including natural language processing (NLP) and explainable AI. His book "Transformers for NLP" explores the application of data science to linguistics, highlighting the unprecedented accuracy of transformers in processing natural language tasks. Rothman explains that NLP involves transforming vast amounts of data into meaningful representations, a process that has evolved significantly over the years.

Up to now, you had all that input, and then you had to get good representations. But there were several models. Like, you would do k means clustering, then you do parsers, then you do recurrent neural networks.

--- Denis Rothman

His work on explainable AI focuses on making complex models understandable, using analogies and practical examples to demystify algorithms and their outputs 3 4.