About the Book

The book focuses on machine learning and systemic machine learning -- a specialized research area in the field of machine learning. It also covers reinforcement learning and advanced developments in this field. There are always difficulties in making machines to learn from experiences. Further always while learning the complete information is not available and it becomes available in bits and pieces over the period of time. The information is even collected and processed from a particular perspective. As per systemic learning there is need to understand impact of decisions and actions on system and that to over the period of time.

About the Author

Dr. Parag Kulkarni has been working in IT industry for last 17 years. He has worked as Research head, principle scientist, chief scientist and was involved in bringing up two startups to speed. He is working as a Chief Scientist at Capsilon INDIA.

Table of Contents:
Preface Acknowledgments About the Author 1 Introduction to Reinforcement and Systemic Machine Learning 1.1. Introduction 1.2. Supervised, Unsupervised, and Semi supervised Machine Learning 1.3. Traditional Learning Methods and History of Machine Learning 1.4. What Is Machine Learning? 1.5. Machine-Learning Problem 1.6. Learning Paradigms 1.7. Machine-Learning Techniques and Paradigms 1.8. What Is Reinforcement Learning? 1.9. Reinforcement Function and Environment Function 1.10. Need of Reinforcement Learning 1.11. Reinforcement Learning and Machine Intelligence 1.12. What is Systemic Learning? 1.13. What is Systemic Machine Learning? 1.14. Challenges in Systemic Machine Learning 1.15. Reinforcement Machine Learning and Systemic Machine Learning 1.16. Case Study Problem Detection in a Vehicle 1.17. Summary 2 Fundamentals of Whole-System, Systemic and Multiperspective Machine Learning 2.1. Introduction 2.2. What Is Systemic Machine Learning? 2.3. Generalized Systemic Machine-Learning Framework 2.4. Multiperspective Decision Making and Multiperspective Learning 2.5. Dynamic and Interactive Decision Making 2.6. The Systemic Learning Framework 2.7. System Analysis 2.8. Case Study: Need of Systemic Learning in the Hospitality Industry 2.9. Summary 3 Reinforcement Learning 3.1. Introduction 3.2. Learning Agents 3.3. Returns and Reward Calculations 3.4. Reinforcement Learning and Adaptive Control 3.5. Dynamic Systems 3.6. Reinforcement Learning and Control 3.7. Markov Property and Markov Decision Process 3.8. Value Functions 3.9. Learning an Optimal Policy (Model-Based and Model-Free Methods) 3.10. Dynamic Programming 3.11. Adaptive Dynamic Programming 3.12. Example: Reinforcement Learning for Boxing Trainer 3.13. Summary 4 Systemic Machine Learning and Model 4.1. Introduction 4.2. A Framework for Systemic Learning 4.3. Capturing the Systemic View 4.4. Mathematical Representation of System Interactions 4.5. Impact Function 4.6. Decision-Impact Analysis 4.7. Summary 5 Inference and Information Integration 5.1. Introduction 5.2. Inference Mechanisms and Need 5.3. Integration of Context and Inference 5.4. Statistical Inference and Induction 5.5. Pure Likelihood Approach 5.6. Bayesian Paradigm and Inference 5.7. Time-Based Inference 5.8. Inference to Build a System View 5.9. Summary 6 Adaptive Learning 6.1. Introduction 6.2. Adaptive Learning and Adaptive Systems 6.3. What Is Adaptive Machine Learning? 6.4. Adaptation and Learning Method Selection Based on Scenario 6.5. Systemic Learning and Adaptive Learning 6.6. Competitive Learning and Adaptive Learning 6.7. Examples 6.8. Summary 7 Multiperspective and Whole-System Learning 7.1. Introduction 7.2. Multiperspective Context Building 7.3. Multiperspective Decision Making and Multiperspective Learning 7.4. Whole-System Learning and Multiperspective Approaches 7.5. Case Study Based on Multiperspective Approach 7.6. Limitations to a Multiperspective Approach 7.7. Summary 8 Incremental Learning and Knowledge Representation 8.1. Introduction 8.2. Why Incremental Learning? 8.3. Learning from What Is Already Learned. . . 8.4. Supervised Incremental Learning 8.5. Incremental Unsupervised Learning and Incremental Clustering 8.6. Semi supervised Incremental Learning 8.7. Incremental and Systemic Learning 8.8. Incremental Closeness Value and Learning Method 8.9. Learning and Decision-Making Model 8.10. Incremental Classification Techniques 8.11. Case Study: Incremental Document Classification 8.12. Summary 9 Knowledge Augmentation: A Machine Learning Perspective 9.1. Introduction 9.2. Brief History and Related Work 9.3. Knowledge Augmentation and Knowledge Elicitation 9.4. Life Cycle of Knowledge 9.5. Incremental Knowledge Representation 9.6. Case-Based Learning and Learning with Reference to Knowledge Loss 9.7. Knowledge Augmentation: Techniques and Methods 9.8. Heuristic Learning 9.9. Systemic Machine Learning and Knowledge Augmentation 9.10. Knowledge Augmentation in Complex Learning Scenarios 9.11. Case Studies 9.12. Summary 10 Building a Learning System 10.1. Introduction 10.2. Systemic Learning System 10.3. Algorithm Selection 10.4. Knowledge Representation 10.5. Designing a Learning System 10.6. Making System to Behave Intelligently 10.7. Example-Based Learning 10.8. Holistic Knowledge Framework and Use of Reinforcement Learning 10.9. Intelligent Agents--Deployment and Knowledge Acquisition and Reuse 10.10. Case-Based Learning: Human Emotion-Detection System 10.11. Holistic View in Complex Decision Problem 10.12. Knowledge Representation and Data Discovery 10.13. Components 10.14. Future of Learning Systems and Intelligent Systems 10.15. Summary Appendix A: Statistical Learning Methods Appendix B: Markov Processes Index