Evolutionary program induction directed by logic grammars.

by Wong Man Leung.Thesis (Ph.D.)--Chinese University of Hong Kong, 1995.Includes bibliographical references (leaves 227-236).List of Figures --- p.iiiList of Tables --- p.viChapter Chapter 1 : --- Introduction --- p.1Chapter 1.1. --- Automatic programming and program induction --- p.1Chapter 1.2. --- Motivation --- p.6Chapter 1.3. --- Contributions of the research --- p.8Chapter 1.4. --- Outline of the thesis --- p.11Chapter Chapter 2 : --- An Overview of Evolutionary Algorithms --- p.13Chapter 2.1. --- Evolutionary algorithms --- p.13Chapter 2.2. --- Genetic Algorithms (GAs) --- p.15Chapter 2.2.1. --- The canonical genetic algorithm --- p.16Chapter 2.2.1.1. --- Selection methods --- p.21Chapter 2.2.1.2. --- Recombination methods --- p.24Chapter 2.2.1.3. --- Inversion and Reordering --- p.27Chapter 2.2.2. --- Implicit parallelism and the building block hypothesis --- p.28Chapter 2.2.3. --- Steady state genetic algorithms --- p.32Chapter 2.2.4. --- Hybrid algorithms --- p.33Chapter 2.3. --- Genetic Programming (GP) --- p.34Chapter 2.3.1. --- Introduction to the traditional GP --- p.34Chapter 2.3.2. --- Automatic Defined Function (ADF) --- p.41Chapter 2.3.3. --- Module Acquisition (MA) --- p.44Chapter 2.3.4. --- Strongly Typed Genetic Programming (STGP) --- p.49Chapter 2.4. --- Evolution Strategies (ES) --- p.50Chapter 2.5. --- Evolutionary Programming (EP) --- p.55Chapter Chapter 3 : --- Inductive Logic Programming --- p.59Chapter 3.1. --- Inductive concept learning --- p.59Chapter 3.2. --- Inductive Logic Programming (ILP) --- p.62Chapter 3.2.1. --- Interactive ILP --- p.64Chapter 3.2.2. --- Empirical ILP --- p.65Chapter 3.3. --- Techniques and methods of ILP --- p.67Chapter Chapter 4 : --- Genetic Logic Programming and Applications --- p.74Chapter 4.1. --- Introduction --- p.74Chapter 4.2. --- Representations of logic programs --- p.76Chapter 4.3. --- Crossover of logic programs --- p.81Chapter 4.4. --- Genetic Logic Programming System (GLPS) --- p.87Chapter 4.5. --- Applications --- p.90Chapter 4.5.1. --- The Winston's arch problem --- p.91Chapter 4.5.2. --- The modified Quinlan's network reachability problem --- p.92Chapter 4.5.3. --- The factorial problem --- p.95Chapter Chapter 5 : --- The logic grammars based genetic programming system (LOGENPRO) --- p.100Chapter 5.1. --- Logic grammars --- p.101Chapter 5.2. --- Representations of programs --- p.103Chapter 5.3. --- Crossover of programs --- p.111Chapter 5.4. --- Mutation of programs --- p.126Chapter 5.5. --- The evolution process of LOGENPRO --- p.130Chapter 5.6. --- Discussion --- p.132Chapter Chapter 6 : --- Applications of LOGENPRO --- p.134Chapter 6.1. --- Learning functional programs --- p.134Chapter 6.1.1. --- Learning S-expressions using LOGENPRO --- p.134Chapter 6.1.2. --- The DOT PRODUCT problem --- p.137Chapter 6.1.2. --- Learning sub-functions using explicit knowledge --- p.143Chapter 6.2. --- Learning logic programs --- p.148Chapter 6.2.1. --- Learning logic programs using LOGENPRO --- p.148Chapter 6.2.2. --- The Winston's arch problem --- p.151Chapter 6.2.3. --- The modified Quinlan's network reachability problem --- p.153Chapter 6.2.4. --- The factorial problem --- p.154Chapter 6.2.5. --- Discussion --- p.155Chapter 6.3. --- Learning programs in C --- p.155Chapter Chapter 7 : --- Knowledge Discovery in Databases --- p.159Chapter 7.1. --- Inducing decision trees using LOGENPRO --- p.160Chapter 7.1.1. --- Decision trees --- p.160Chapter 7.1.2. --- Representing decision trees as S-expressions --- p.164Chapter 7.1.3. --- The credit screening problem --- p.166Chapter 7.1.4. --- The experiment --- p.168Chapter 7.2. --- Learning logic program from imperfect data --- p.174Chapter 7.2.1. --- The chess endgame problem --- p.177Chapter 7.2.2. --- The setup of experiments --- p.178Chapter 7.2.3. --- Comparison of LOGENPRO with FOIL --- p.180Chapter 7.2.4. --- Comparison of LOGENPRO with BEAM-FOIL --- p.182Chapter 7.2.5. --- Comparison of LOGENPRO with mFOILl --- p.183Chapter 7.2.6. --- Comparison of LOGENPRO with mFOIL2 --- p.184Chapter 7.2.7. --- Comparison of LOGENPRO with mFOIL3 --- p.185Chapter 7.2.8. --- Comparison of LOGENPRO with mFOIL4 --- p.186Chapter 7.2.9. --- Comparison of LOGENPRO with mFOIL5 --- p.187Chapter 7.2.10. --- Discussion --- p.188Chapter 7.3. --- Learning programs in Fuzzy Prolog --- p.189Chapter Chapter 8 : --- An Adaptive Inductive Logic Programming System --- p.192Chapter 8.1. --- Adaptive Inductive Logic Programming --- p.192Chapter 8.2. --- A generic top-down ILP algorithm --- p.196Chapter 8.3. --- Inducing procedural search biases --- p.200Chapter 8.3.1. --- The evolution process --- p.201Chapter 8.3.2. --- The experimentation setup --- p.202Chapter 8.3.3. --- Fitness calculation --- p.203Chapter 8.4. --- Experimentation and evaluations --- p.204Chapter 8.4.1. --- The member predicate --- p.205Chapter 8.4.2. --- The member predicate in a noisy environment --- p.205Chapter 8.4.3. --- The multiply predicate --- p.206Chapter 8.4.4. --- The uncle predicate --- p.207Chapter 8.5. --- Discussion --- p.208Chapter Chapter 9 : --- Conclusion and Future Work --- p.210Chapter 9.1. --- Conclusion --- p.210Chapter 9.2. --- Future work --- p.217Chapter 9.2.1. --- Applying LOGENPRO to discover knowledge from databases --- p.217Chapter 9.2.2. --- Learning recursive programs --- p.218Chapter 9.2.3. --- Applying LOGENPRO in engineering design --- p.220Chapter 9.2.4. --- Exploiting parallelism of evolutionary algorithms --- p.222Reference --- p.227Appendix A --- p.23