Calculus of Thought

Preface

A Personal Perspective

Chapter 1. Calculus Ratiocinator

Abstract

1 A Fundamental Problem with the Widely Used Methods

2 Ensemble Models and Cognitive Processing in Playing Jeopardy

3 The Brain's Explicit and Implicit Learning

4 Two Distinct Modeling Cultures and Machine Intelligence

5 Logistic Regression and the Calculus Ratiocinator Problem

Chapter 2. Most Likely Inference

Abstract

1 The Jaynes Maximum Entropy Principle

2 Maximum Entropy and Standard Maximum Likelihood Logistic Regression

3 Discrete Choice, Logit Error, and Correlated Observations

4 RELR and the Logit Error

5 RELR and the Jaynes Principle

Chapter 3. Probability Learning and Memory

Abstract

1 Bayesian Online Learning and Memory

2 Most Probable Features

3 Implicit RELR

4 Explicit RELR

Chapter 4. Causal Reasoning

Abstract

1 Propensity Score Matching

2 RELR's Outcome Score Matching

3 An Example of RELR's Causal Reasoning

4 Comparison to Other Bayesian and Causal Methods

Chapter 5. Neural Calculus

Abstract

1 RELR as a Neural Computational Model

2 RELR and Neural Dynamics

3 Small Samples in Neural Learning

4 What about Artificial Neural Networks?

Chapter 6. Oscillating Neural Synchrony

Abstract

1 The EEG and Neural Synchrony

2 Neural Synchrony, Parsimony, and Grandmother Cells

3 Gestalt Pragnanz and Oscillating Neural Synchrony

4 RELR and Spike-Timing-Dependent Plasticity

5 Attention and Neural Synchrony

6 Metrical Rhythm in Oscillating Neural Synchrony

7 Higher Frequency Gamma Oscillations

Chapter 7. Alzheimer's and Mind–Brain Problems

Abstract

1 Neuroplasticity Selection in Development and Aging

2 Brain and Cognitive Changes in Very Early Alzheimer's Disease

3 A RELR Model of Recent Episodic and Semantic Memory

4 What Causes the Medial Temporal Lobe Disturbance in Early Alzheimer's?

5 The Mind–Brain Problem

Chapter 8. Let Us Calculate

Abstract

1 Human Decision Bias and the Calculus Ratiocinator

2 When the Experts are Wrong

3 When Predictive Models Crash

4 The Promise of Cognitive Machines

Appendix

A1 RELR Maximum Entropy Formulation

A2 Derivation of RELR Logit from Errors-in-Variables Considerations

A3 Methodology for Pew 2004 Election Weekend Model Study

A4 Derivation of Posterior Probabilities in RELR's Sequential Online Learning

A5 Chain Rule Derivation of Explicit RELR Feature Importance

A6 Further Details on the Explicit RELR Low Birth Weight Model in Chapter 3

A7 Zero Intercepts in Perfectly Balanced Stratified Samples

A8 Detailed Steps in RELR's Causal Machine Learning Method

Notes and References

Index