Description
In Build a Reasoning Model, Sebastian Raschka tackles one of the hottest topics in the world of AI: reasoning models (such as OpenAI’s o1 series). Instead of using ready-made models, he teaches the reader how to use a pre-built language model to enhance and implement step-by-step thinking capabilities and solve complex logical problems.
The tutorials are very practical and focus on the idea that real learning only comes through doing. The book focuses on techniques that allow a model to explore different paths and correct its own mistakes before providing a final answer, a process that makes AI much more accurate in areas like math and coding.
Book Features
- Learn to implement core reasoning improvements for large language models (LLM).
- Teaching model evaluation methods based on human judgment criteria and benchmarks.
- Focus on improving reasoning without the need to constantly update heavy model weights.
- Learning how to work with KV cache and techniques for optimizing text production speed.
- Using reinforcement learning (RL) approaches to improve problem-solving ability.
- Providing executable code on common and home hardware.
Book specifications
- Publisher: MANNING
- Instructor/Author: Sebastian Raschka
- Number of pages: 363
- Number of chapters: 6
- Format: PDF
Headlines
Pictures
User Guide
Extract the file and run it with the appropriate software.
Download link
File(s) password: www.downloadly.ir
File size
20.3 MB
