Learning Functions from Imperfect Positive Data
Abstract:
Abstract. The Bayesian framework of learning from positive noise-free examples derived by Muggleton [12] is extended to learning functional hypotheses from positive examples containing normally distributed noise in the outputs. The method subsumes a type of distance based learning as a special case. We also present an effective method of outlieridentification which may significantly improve the predictive accuracy of the final multi-clause hypothesis if it is constructed by a clause-by-clause covering algorithm as e.g. in Progol or Aleph. Our method is implemented in Aleph and tested on two experiments, one of which concerns numeric functions while the other treats non-numeric discrete data where the normal distribution is taken as an approximation of the discrete distribution of noise. 1
Citations
| 486 | Inverse entailment and Progol – Muggleton - 1995 |
| 181 | Theory of Optimal Experiments – Fedorov - 1972 |
| 68 | Learning from positive data – Muggleton - 1996 |
| 62 | Induction of first-order decision lists : Results on learning the past tense of english verbs – Mooney, Califf - 1995 |
| 60 | Relational instance-based learning – Emde, Wettschereck - 1996 |
| 27 | Distance between Herbrand interpretations: a measure for approximations to a target concept – Nienhuys-Cheng - 1997 |
| 18 | A learnability model for universal representations – Muggleton, Page - 1994 |
| 14 | Handling imperfect data in inductive logic programming – Lavrac, Dzeroski, et al. - 1996 |
| 8 | Analogical prediction – Muggleton, Bain - 1999 |
| 4 | Decomposing probability distributions on structured individuals – Flach, Lachiche - 2000 |
| 3 | Instance based function learning – Ramon, Raedt - 1999 |
