This book constitutes the thoroughly refereed post-conference proceedings of the 17th International Conference on Inductive Logic Programming, ILP 2007, held in Corvallis, OR, USA, in June 2007 in conjunction with ICML 2007, the International Conference on Machine Learning.
The 15 revised full papers and 11 revised short papers presented together with 2 invited lectures were carefully reviewed and selected from 38 initial submissions. The papers present original results on all aspects of learning in logic, as well as multi-relational learning and data mining, statistical relational learning, graph and tree mining, relational reinforcement learning, and learning in other non-propositional knowledge representation frameworks. Thus all current topics in inductive logic programming, ranging from theoretical and methodological issues to advanced applications in various areas are covered.
Author(s): Hendrik Blockeel, Jan Ramon, Jude Shavlik, Prasad Tadepalli
Series: Lecture Notes in Computer Science - Lecture Notes Artificial Intelligence
Publisher: Springer
Year: 2008
Language: English
Pages: 317
front-matter......Page 1
Motivations......Page 11
Overview of Methods......Page 12
Introduction......Page 14
Why Causal Models Are Useful......Page 15
NASD......Page 17
Available Data......Page 18
Current Practice......Page 19
Experimental Design......Page 20
Joint Modeling......Page 21
Quasi-experimental Design......Page 23
Limitations of Current Approaches......Page 25
Research Opportunities......Page 26
Timeliness......Page 28
Risks and Benefits......Page 29
References......Page 30
Learning Probabilistic Logic Models from Probabilistic Examples (Extended Abstract)......Page 32
Learning Directed Probabilistic Logical Models Using Ordering-Search......Page 34
Learning to Assign Degrees of Belief in Relational Domains......Page 35
Bias/Variance Analysis for Relational Domains......Page 37
Introduction......Page 39
A Family of Semi-distances for Instances......Page 40
Discussion......Page 41
Extensions......Page 42
Optimization......Page 43
Experiments on Similarity Search......Page 45
Conclusions and Ongoing Work......Page 47
Introduction......Page 49
Method......Page 50
Experiments......Page 51
Summary and Future Work......Page 57
Introduction......Page 59
Statistical Structural Software Testing......Page 60
SSST and Supervised Learning......Page 62
Principle......Page 63
Init Module......Page 65
Constrained Exploration Module......Page 66
Experimental Setting......Page 67
Experimental Results......Page 68
Conclusion and Perspectives......Page 70
Introduction......Page 73
Inductive Process Modeling......Page 74
Learning Bias by Inductive Logic Programming......Page 76
Empirical Evaluation......Page 78
Method......Page 79
Selecting a Performance Threshold......Page 80
Evaluating the Generalization Performance......Page 81
Semantic Analysis of the Induced Constraints......Page 82
General Discussion......Page 84
Conclusion......Page 86
Introduction......Page 88
The Trie Data Structure......Page 89
Trieing MDIE......Page 91
Trie the Real World......Page 94
Experiments and Results......Page 95
Conclusions......Page 96
Introduction......Page 98
Relational Reinforcement Learning and the Blocks World......Page 99
Transfer Learning......Page 100
Relational Options......Page 101
Relational Skill Learning......Page 102
Experimental Evaluation......Page 104
Conclusions and Further Work......Page 106
Introduction......Page 108
Problem Settings......Page 109
Score Propagation......Page 111
Related Work......Page 112
Class Overlap......Page 113
ROC Analysis......Page 115
Experiments with the CoRA Data......Page 116
Discussion and Future Work......Page 119
Introduction......Page 122
State of the Art......Page 123
When MI Learning Meets Linear Programming......Page 124
Order Parameters and Experimental Setting......Page 126
Experiments......Page 127
LPP Satisfiability Landscape......Page 128
Generalization Error Landscape......Page 129
Conclusion and Perspectives......Page 130
Introduction......Page 132
Gleaning Clauses......Page 133
Creating Features......Page 134
Learning to Predict Scores......Page 135
Experimental Results......Page 137
Related and Future Work......Page 139
Introduction......Page 142
A Representation for Process Traces and Models......Page 143
Learning ICs Theories......Page 145
Experiments......Page 148
Related Works......Page 153
Conclusions and Future Works......Page 154
Introduction......Page 157
Description Logics......Page 158
Learning in Description Logics Using Refinement Operators......Page 159
Designing a Refinement Operator......Page 161
Completeness of the Operator......Page 162
Achieving Properness......Page 163
Redundancy Elimination......Page 164
Creating a Full Learning Algorithm......Page 165
Preliminary Evaluation......Page 167
Related Work......Page 168
Conclusions and Further Work......Page 169
Introduction......Page 171
Description Logics......Page 173
Learning in Description Logics Using Refinement Operators......Page 174
Analysing the Properties of Refinement Operators......Page 176
Related Work......Page 181
Conclusions......Page 183
Introduction......Page 185
Decision-Theoretic Assistance......Page 187
Relational Hierarchical Policies......Page 188
Goal Estimation......Page 191
Action Selection......Page 193
Doorman Domain......Page 194
Kitchen Domain......Page 196
Related Work......Page 198
Conclusions and Future Work......Page 199
Introduction......Page 201
Modeling ILP's Search Space with Bayesian Networks......Page 202
Training the Model......Page 203
Using the Model to Guide Search......Page 205
Directed-Search Experiments......Page 206
Conclusions and Future Work......Page 208
Introduction......Page 210
Stochastic Search......Page 211
First-Order Logic Theory Revision......Page 212
Stochastic First-Order Logic Theory Revision......Page 213
Stochastic Local Search for Antecedents......Page 214
Stochastic Local Search for Revisions......Page 215
Experimental Results......Page 216
Conclusions......Page 219
Introduction......Page 221
Feature Definitions Using Inductive Logic Programming......Page 224
Aims......Page 226
Materials......Page 227
Method......Page 229
Results......Page 230
Concluding Remarks......Page 232
Introduction......Page 235
Mode Declarations......Page 236
MDIE (Mode Directed Inverse Entailment)......Page 237
HAIL (Hybrid Abductive Inductive Learning)......Page 238
SOLAR (SOL Resolution for Advanced Reasoning)......Page 239
Motivating Example: Fluid Modelling......Page 240
Full Clausal Hybrid Abductive Inductive Learning......Page 243
Related Work......Page 246
Conclusions......Page 247
Introduction......Page 249
Graph Pattern......Page 251
Block Preserving Outerplanar Graph Patterns and Block Tree Patterns......Page 252
Matching Algorithm for Block Preserving Outerplanar Graph Patterns......Page 255
Pattern Enumeration Algorithm for Frequent BPO Graph Pattern Problem......Page 257
Experimental Result......Page 261
Conclusion and Future Works......Page 262
Introduction......Page 264
Reinforcement Learning in RoboCup......Page 265
Related Work in Transfer Learning......Page 267
Executing a Relational Macro......Page 268
Structure Learning......Page 269
Ruleset Learning......Page 271
Transferring a Relational Macro......Page 273
Experimental Results......Page 275
Conclusions and Future Work......Page 277
Introduction......Page 279
Computing Heuristics from the Ensemble......Page 281
Generation of Candidate Test Queries......Page 282
Computing the Optimal Split......Page 283
Empirical Evaluation......Page 285
Conclusions and Future Work......Page 287
Introduction......Page 290
Background......Page 291
Building World Models......Page 292
Algorithm Overview......Page 293
Preimage Selection......Page 294
Building the MDP......Page 295
The RL Learning Cycle......Page 296
Domains......Page 297
Learning Algorithms......Page 298
Related Work......Page 299
Conclusions and Future Work......Page 300
References......Page 301
Introduction......Page 302
Representation of Individuals......Page 303
Representation of Features......Page 305
Structured Feature Selection......Page 307
Node Selection......Page 309
The Precision/Recall-Driven Decision-Tree Algorithm......Page 310
The Dataset......Page 312
Experimental Results......Page 314
Conclusion......Page 316
back-matter......Page 317
