Description
Mathematics Behind Large Language Models and Transformers is a course on the fundamental mathematics that underpin advanced AI systems such as GPT and BERT, published by Udemy Online Academy. The course explores linear algebra, calculus, probability, and optimization techniques that underpin transformer models, including attention mechanisms, token embedding, and gradient descent optimization. The course explores the complex mathematical algorithms that allow these complex models to process, understand, and produce human-like text. Starting with tokenization, students will learn how raw text is transformed into a format that can be understood by the models through techniques such as the WordPiece algorithm.
By the end of this course, individuals will not only have mastered the theoretical foundations of transformers, but also gain practical insight into their operation and application. This course will prepare you for innovation and excellence in machine learning, placing you among the top engineers and researchers in artificial intelligence. This course is recommended for ambitious learners aiming to reach the top ranks of the programming world and those who want to gain a deep understanding of transformers, the advanced technology behind large language models.
What you will learn in Mathematics Behind Large Language Models and Transformers:
- The Mathematics Behind Large Language Models
- Query, Value, and Key Matrix
- Masked Language Modeling
- The Nature of Sine and Cosine Functions in Positional Encodings
- Word Embeddings
- Matrix Multiplication
- And…
Course specifications
- Publisher : Udemy
- Teacher : Patrik Szepesi
- Language: English
- Level : All Levels
- Lectures : 48
- Duration : 6 hours and 56 minutes
Course topics
Mathematics Behind Large Language Models and Transformers Prerequisites
Basic HS math(linear algebra)
Pictures
Mathematics Behind Large Language Models and Transformers introduction video
Installation guide
After Extract, watch with your favorite Player.
Subtitle: English
Quality: 1080p
The 2026/2 version has increased the number of lessons by 19 and the duration increased by 2 hours 14 minutes compared to 2024/5.
Download link
File password (s): www.downloadly.ir
Size
2.67 GB
