Causality, Correlation And Artificial Intelligence For Rational Decision Making

✍ Scribed by Tshilidzi Marwala

• Publisher: World Scientific Publishing
• Year: 2015
• Tongue: English
• Leaves: 207
• Category: Library

✦ Synopsis

Causality has been a subject of study for a long time. Often causality is confused with correlation. Human intuition has evolved such that it has learned to identify causality through correlation. In this book, four main themes are considered and these are causality, correlation, artificial intelligence and decision making. A correlation machine is defined and built using multi-layer perceptron network, principal component analysis, Gaussian Mixture models, genetic algorithms, expectation maximization technique, simulated annealing and particle swarm optimization. Furthermore, a causal machine is defined and built using multi-layer perceptron, radial basis function, Bayesian statistics and Hybrid Monte Carlo methods. Both these machines are used to build a Granger non-linear causality model. In addition, the Neyman-Rubin, Pearl and Granger causal models are studied and are unified. The automatic relevance determination is also applied to extend Granger causality framework to the non-linear domain. The concept of rational decision making is studied, and the theory of flexibly-bounded rationality is used to extend the theory of bounded rationality within the principle of the indivisibility of rationality. The theory of the marginalization of irrationality for decision making is also introduced to deal with satisficing within irrational conditions. The methods proposed are applied in biomedical engineering, condition monitoring and for modelling interstate conflict.

✦ Table of Contents

Contents
Preface
Acknowledgments
1. Introduction to Artificial Intelligence based Decision Making
1.1 Introduction
1.2 Correlation
1.2.1 What is correlation?
1.2.2 Correlation function
1.3 Causality
1.3.1 What is causality?
1.3.2 Theories of causality
1.3.3 What is a causal function?
1.3.4 How to detect causation?
1.4 Introduction to Artificial Intelligence
1.4.1 Neural networks
1.4.2 Hopfield networks
1.4.3 Genetic algorithm
1.4.4 Particle swarm optimization
1.4.5 Simulated annealing
1.5 Rational Decision Making
1.6 Summary and Outline of the Book
1.7 Conclusions
References
2. What is a Correlation Machine?
2.1 Introduction
2.2 Correlation Machines
2.2.1 Auto-associative memory network
2.2.2 Principal component analysis
2.2.3 Expectation maximization algorithm
2.3 Genetic Algorithm
2.3.1 Initialization
2.3.2 Crossover
2.3.3 Mutation
2.3.4 Reproduction
2.3.5 Termination
2.4 Multi-layer Perceptron
2.5 Experimental Comparison
2.6 Conclusions
References
3. What is a Causal Machine?
3.1 Introduction
3.2 Induction, Deduction, and Abduction
3.3 What is Causality?
3.4 Multi-layer Perceptron Causal Machine
3.4.1 The architecture of the MLP causal machine
3.4.2 Interstate conflict
3.5 Radial Basis Function Causal Machine
3.5.1 Theoretical foundation
3.5.2 Applications to condition monitoring
3.6 Fuzzy Inference System Causal Machine
3.6.1 Theoretical foundation
3.6.2 Application to a steam generator
3.7 Conclusions
References
4. Correlation Machines Using Optimization Methods
4.1 Introduction
4.2 Multi-layer Perceptron Neural Network
4.3 Missing Data Estimation Technique
4.4 Genetic Algorithms
4.4.1 Initialization
4.4.2 Crossover
4.4.3 Mutation
4.4.4 Reproduction
4.4.5 Termination
4.5 Particle Swarm Optimization
4.6 Simulated Annealing
4.6.1 SA parameters
4.6.2 Transition probabilities
4.6.3 Monte Carlo method
4.6.4 Markov Chain Monte Carlo
4.6.5 Acceptance probability function: Metropolis algorithm
4.6.6 Cooling schedule
4.7 Missing Data Estimation: Case Studies
4.7.1 Mechanical system
4.7.2 Modeling of beer tasting
4.8 Conclusions
References
5. Neural Networks for Modeling Granger Causality
5.1 Introduction
5.2 Granger Causality
5.3 Multi-layer Perceptron for Granger Causality
5.3.1 Bayesian statistics
5.3.2 Hybrid Monte Carlo (HMC)
5.4 RBF for Granger Causality
5.4.1 The k-means
5.4.2 Pseudo-inverse methods
5.5 Example: Mackey–Glass System
5.6 Conclusions
References
6. Rubin, Pearl and Granger Causality Models: A Unified View
6.1 Introduction
6.2 Neyman-Rubin Causal Model
6.2.1 Missing data mechanism
6.2.2 Missing data imputation methods
6.3 Pearl Causality
6.3.1 Directed acyclic graph
6.3.2 Associations between variables
6.3.3 d-separation
6.3.4 Back-door adjustment
6.3.5 Front-door adjustment
6.3.6 Rules for do-calculus
6.3.7 Pearl inferred causation algorithm
6.3.8 Examples of using do-calculus
6.4 Granger Causality
6.5 Comparison: Neyman-Rubin, Pearl and Granger Causality
6.6 Conclusions
References
7. Causal, Correlation and Automatic Relevance Determination Machines for Granger Causality
7.1 Introduction
7.2 Causal Machine to Granger Causality
7.2.1 Multi-layer perceptron
7.2.2 Scaled conjugate gradient method
7.3 Correlation Machine to Granger Causality
7.3.1 Auto-associative network for missing data estimation
7.3.2 Nelder-Mead simplex optimization method
7.3.3 Granger causality
7.4 Automatic Relevance Determination for Granger Causality
7.5 Experimental Investigation: Mackey–Glass Time-Delay Differential Equation
7.6 Conclusions
References
8. Flexibly-bounded Rationality
8.1 Introduction
8.2 Rational Decision Making: A Causal Approach
8.3 Rational Decision Making Process
8.4 Bounded-Rational Decision Making
8.5 Flexibly-bounded Rational Decision Making
8.5.1 Advanced information processing
8.5.2 Missing data estimation
8.5.3 Intelligent machines
8.6 Experimental Investigations
8.6.1 Condition monitoring
8.6.2 HIV modeling
8.7 Conclusions
References
9. Marginalization of Irrationality in Decision Making
9.1 Introduction
9.2 Rational Decision Making
9.3 What is Irrationality?
9.4 Marginalization of Irrationality Theory
9.5 Irrational Decision Making and the Theory of Marginalization of Irrationality in Decision Making
9.6 Application of the Marginalization of Irrationality Theory for Breast Cancer Diagnosis
9.6.1 MLP
9.6.2 RBF
9.6.3 Auto-associative neural network based on the MLP
9.6.4 Auto-associative network based on the RBF
9.7 Conclusions
References
10. Conclusions and FurtherWork
10.1 Introduction
10.2 Way Forward
References
Index