Machine Learning Systems: Design and Implementation

《Machine Learning Systems: Design and Implementation》

Coming Soon!

Be the world's first open source book that comprehensively introduces the knowledge of machine learning systems

Star

Core Author

Luo Mai

University of Edinburgh

Hao Dong

Peking University, Peng Cheng Laboratory

Xuefeng Jin

MindSpore Architect

Zhiliang Gan

MindSpore Architect

Chapter Author

Cheng Lai

Peng Cheng Laboratory

Jiarong Han

Peng Cheng Laboratory

Yao Fu

University of Edinburgh

Xiulong Yuan

Tsinghua University

Zihan Ding

Princeton University

Jiankai Sun

Stanford University

Peiyuan Liao

Carnegie Mellon University

Wenteng Liang

Beijing University of Posts and Telecommunications

Jie Ren

University of Edinburgh

Hanchen Wang

University of Cambridge

Pei Mu

University of Edinburgh

Qinghua Zhang

Huawei Engineer

Zhibo Liang

Huawei Engineer

Jianfeng Yu

Huawei Engineer

Jinjin Chu

Huawei Engineer

Fubi Cai

Huawei Engineer

Renwei Zhang

Huawei Engineer

Chao Liu

Huawei Engineer

Gang Chen

Huawei Engineer

Mingqi Li

Huawei Engineer

Gangqiang Han

Huawei Engineer

Yehui Tang

Huawei Engineer

Zhiqiang Zhai

Huawei Engineer

Tiancheng Wu

Huawei Engineer

Xiaohui Li

Huawei Engineer

Haoyang Li

Huawei Engineer

Zhipeng Tan

Huawei Engineer

Shanni Li

Huawei Engineer

Zhijian Guo

Huawei Engineer

Thanks to contributors from the community

Thanks to readers who provided valuable comments for this book

Contributions to this book are welcome

Table of contents

• 1. Preface
  • 1.1. Background
  • 1.2. Beginning
  • 1.3. Community Building
• 2. Introduction
  • 2.1. Machine Learning Applications
  • 2.2. Design Objectives of Machine Learning Frameworks
  • 2.3. Machine Learning Framework Architecture
  • 2.4. Application Scenarios of Machine Learning Systems
  • 2.5. Book Organization and Intended Audience
• 3. Part I Framework Design
• 4. Programming Model
  • 4.1. Overview
  • 4.2. Machine Learning Workflow
  • 4.3. Neural Network Programming
  • 4.4. Functional Programming
  • 4.5. Bridging Python and C/C++ Functions
  • 4.6. Chapter Summary
• 5. AI Compiler Frontend
  • 5.1. Overview of AI Compilers
  • 5.2. Overview of AI Compiler Frontends
  • 5.3. Intermediate Representation
  • 5.4. Automatic Differentiation
  • 5.5. Type Systems and Static Analysis
  • 5.6. Frontend Compilation Optimization
  • 5.7. Chapter Summary
  • 5.8. Further Reading
• 6. AI Compiler Backend
  • 6.1. Overview
  • 6.2. Graph Optimization
  • 6.3. Operator Selection
  • 6.4. Memory Allocation
  • 6.5. Operator Compiler
  • 6.6. Chapter Summary
  • 6.7. Further Reading
• 7. Hardware Accelerator
  • 7.1. Overview
  • 7.2. Components of Hardware Accelerators
  • 7.3. Programming Methods
  • 7.4. Performance Optimization Methods
  • 7.5. Chapter Summary
• 8. Distributed Training
  • 8.1. Overview
  • 8.2. Parallelism Methods
  • 8.3. Pipeline Parallelism with Micro-Batching
  • 8.4. Architecture of Machine Learning Clusters
  • 8.5. Collective Communication
  • 8.6. Parameter Server
  • 8.7. Federated Learning
  • 8.8. Training Large Language Models
  • 8.9. Chapter Summary
  • 8.10. Further Reading
• 9. Model Deployment
  • 9.1. Overview
  • 9.2. Conversion to Inference Model and Model Optimization
  • 9.3. Model Compression
  • 9.4. Advanced Efficient Techniques
  • 9.5. Model Inference
  • 9.6. Security Protection of Models
  • 9.7. Chapter Summary
  • 9.8. Further Reading
• 10. Part II Application Scenarios
• 11. Recommender System
  • 11.1. Overview
  • 11.2. System Components
  • 11.3. Recommendation Pipeline
  • 11.4. Model Update
  • 11.5. Supporting Real-time Machine Learning
  • 11.6. Chapter Summary
  • 11.7. Further Reading
• 12. Reinforcement Learning System
  • 12.1. Overview
  • 12.2. Introduction to Reinforcement Learning
  • 12.3. Single-Node Reinforcement Learning System
  • 12.4. Distributed Reinforcement Learning System
  • 12.5. Multi-agent Reinforcement Learning
  • 12.6. Multi-agent Reinforcement Learning System
  • 12.7. Chapter Summary
• 13. Robotic System
  • 13.1. Overview of Robotic Systems
  • 13.2. Robot Operating System
  • 13.3. Case Study: Using ROS
  • 13.4. Modern Robot Learning
  • 13.5. Chapter Summary