English | MP4 | AVC 1920×1080 | AAC 44KHz 2ch | 32 Lessons (4h 55m) | 688 MB
Unlock the secrets behind transformers like GPT and BERT. Learn tokenization, attention mechanisms, positional encodings, and embeddings to build and innovate with advanced AI. Excel in the field of machine learning and become a top-tier AI expert.
Discover the mathematical foundations behind transformers like GPT and BERT. From tokenization to attention mechanisms, gain a deep understanding of the key algorithms driving AI advancements. Elevate your skills to innovate and lead in machine learning.
What you’ll learn
- How tokenization transforms text into model-readable data
- The inner workings of attention mechanisms in transformers
- How positional encodings preserve sequence data in AI models
- The role of matrices in encoding and processing language
- Building dense word representations with multi-dimensional embeddings
- Differences between bidirectional and masked language models
- Practical applications of dot products and vector mathematics in AI
- How transformers process, understand, and generate human-like text
Table of Contents
1 Advanced AI: LLMs Explained with Math
2 Creating Our Optional Experiment Notebook – Part 1
3 Creating Our Optional Experiment Notebook – Part 2
4 Encoding Categorical Labels to Numeric Values
5 Understanding the Tokenization Vocabulary
6 Encoding Tokens
7 Practical Example of Tokenization and Encoding
8 DistilBert vs. Bert Differences
9 Embeddings In A Continuous Vector Space
10 Introduction To Positional Encodings
11 Positional Encodings – Part 1
12 Positional Encodings – Part 2 (Even and Odd Indices)
13 Why Use Sine and Cosine Functions
14 Understanding the Nature of Sine and Cosine Functions
15 Visualizing Positional Encodings in Sine and Cosine Graphs
16 Solving the Equations to Get the Values for Positional Encodings
17 Introduction to Attention Mechanism
18 Query, Key and Value Matrix
19 Getting Started with Our Step by Step Attention Calculation
20 Calculating Key Vectors
21 Query Matrix Introduction
22 Calculating Raw Attention Scores
23 Understanding the Mathematics Behind Dot Products and Vector Alignment
24 Visualizing Raw Attention Scores in 2D
25 Converting Raw Attention Scores to Probability Distributions with Softmax
26 Normalization
27 Understanding the Value Matrix and Value Vector
28 Calculating the Final Context Aware Rich Representation for the Word “River”
29 Understanding the Output
30 Understanding Multi Head Attention
31 Multi Head Attention Example and Subsequent Layers
32 Masked Language Learning
Resolve the captcha to access the links!