Results 1 
6 of
6
Learning Stochastic Logic Programs
, 2000
"... Stochastic Logic Programs (SLPs) have been shown to be a generalisation of Hidden Markov Models (HMMs), stochastic contextfree grammars, and directed Bayes' nets. A stochastic logic program consists of a set of labelled clauses p:C where p is in the interval [0,1] and C is a firstorder range ..."
Abstract

Cited by 1057 (71 self)
 Add to MetaCart
Stochastic Logic Programs (SLPs) have been shown to be a generalisation of Hidden Markov Models (HMMs), stochastic contextfree grammars, and directed Bayes' nets. A stochastic logic program consists of a set of labelled clauses p:C where p is in the interval [0,1] and C is a firstorder rangerestricted definite clause. This paper summarises the syntax, distributional semantics and proof techniques for SLPs and then discusses how a standard Inductive Logic Programming (ILP) system, Progol, has been modied to support learning of SLPs. The resulting system 1) nds an SLP with uniform probability labels on each definition and nearmaximal Bayes posterior probability and then 2) alters the probability labels to further increase the posterior probability. Stage 1) is implemented within CProgol4.5, which differs from previous versions of Progol by allowing userdefined evaluation functions written in Prolog. It is shown that maximising the Bayesian posterior function involves nding SLPs with short derivations of the examples. Search pruning with the Bayesian evaluation function is carried out in the same way as in previous versions of CProgol. The system is demonstrated with worked examples involving the learning of probability distributions over sequences as well as the learning of simple forms of uncertain knowledge.
Rule Induction with CN2: Some Recent Improvements
, 1991
"... The CN2 algorithm induces an ordered list of classification rules from examples using entropy as its search heuristic. In this short paper, we describe two improvements to this algorithm. Firstly, we present the use of the Laplacian error estimate as an alternative evaluation function and secondly, ..."
Abstract

Cited by 324 (2 self)
 Add to MetaCart
The CN2 algorithm induces an ordered list of classification rules from examples using entropy as its search heuristic. In this short paper, we describe two improvements to this algorithm. Firstly, we present the use of the Laplacian error estimate as an alternative evaluation function and secondly, we show how unordered as well as ordered rules can be generated. We experimentally demonstrate significantly improved performances resulting from these changes, thus enhancing the usefulness of CN2 as an inductive tool. Comparisons with Quinlan's C4.5 are also made. Keywords: learning, rule induction, CN2, Laplace, noise 1 Introduction Rule induction from examples has established itself as a basic component of many machine learning systems, and has been the first ML technology to deliver commercially successful applications (eg. the systems GASOIL [Slocombe et al., 1986], BMT [HayesMichie, 1990], and in process control [Leech, 1986]). The continuing development of inductive techniques is t...
Building Symbolic Representations of Intuitive Realtime Skills from Performance Data
 In Machine Intelligence 13
, 1994
"... Realtime control skills are ordinarily tacit  their possessors cannot explicitly communicate them. But given sufficient sampling of a trained expert's inputoutput behaviour, machine learning programs have been found capable of constructing rules which, when run as programs, deliver behaviours ..."
Abstract

Cited by 15 (6 self)
 Add to MetaCart
Realtime control skills are ordinarily tacit  their possessors cannot explicitly communicate them. But given sufficient sampling of a trained expert's inputoutput behaviour, machine learning programs have been found capable of constructing rules which, when run as programs, deliver behaviours similar to those of the original exemplars. These `clones' are in effect symbolic representations of subcognitive behaviours. After validation on simple polebalancing tasks, the principles have been successfully generalized in flightsimulator experiments, both by Sammut and others at UNSW, and by Camacho at the Turing Institute. A flight plan switches control through a sequence of logically concurrent sets of reactive behaviours. Each set can be thought of as a committee of subpilots who are respectively specialized for rudder, elevators, rollers, thrust, etc. The chairman (the flight plan) knows only the mission sequence, and how to recognize the onset of each stage. This treatment is ess...
A Model of Argumentation and Its Application in a Cooperative Expert System
, 1991
"... This thesis is about a particular type of problemsolving, based on the discussion which occurs between experts when addressing difficult problems. We term this activity argumentation. Argumentation is characterised by its interactive or social nature, and is based on the construction, criticism, ju ..."
Abstract

Cited by 4 (0 self)
 Add to MetaCart
This thesis is about a particular type of problemsolving, based on the discussion which occurs between experts when addressing difficult problems. We term this activity argumentation. Argumentation is characterised by its interactive or social nature, and is based on the construction, criticism, justification and modification of arguments. In this thesis we present a model of argumentation, and evaluate an implementation's utility for decision support where the system is placed in the role of a colleague to its users. Keywords: argumentation, cooperative systems, expert systems, Toulmin, casebased reasoning, knowledge acquisition, problemsolving. iii Acknowledgements I am greatly indebted to Tim Niblett for the many useful discussions concerning this thesis work, from which many of the ideas presented here have arisen. I am also particularly grateful to our oil exploration experts Mike Whyatt, Nigel Capon and Dave Rhodes of Enterprise Oil plc for their enthusiasm, ideas about and i...
Applications of Machine Learning: Notes from the Panel Members
, 1991
"... Introduction Machine learning (ML) is devoted to the study and computer modelling of learning processes in their multiple manifestations. Although ML research started with the advent of computers, it is only relatively recently that its results have left the research laboratories and found their wa ..."
Abstract

Cited by 1 (0 self)
 Add to MetaCart
Introduction Machine learning (ML) is devoted to the study and computer modelling of learning processes in their multiple manifestations. Although ML research started with the advent of computers, it is only relatively recently that its results have left the research laboratories and found their way into realworld applications. The motivation for applying ML techniques to realworld tasks is strong: the problems of manually engineering a knowledge base are now well known, and ML offers a technology for assisting in this task; there is vast potential for automatically discovering new knowledge in the recent explosion of available online databases, too large for humans to manually sift through; and the ability of computers to automatically adapt to changing expertise would offer huge benefits for the maintenance and evolution of expert systems. Despite this, the success of ML applications varies tremendously. There are some spectacular successes to its credit, but also the num
Applications of Machine Learning: Notes from the Panel Members
, 1991
"... Introduction Machine learning (ML) is devoted to the study and computer modelling of learning processes in their multiple manifestations. Although ML research started with the advent of computers, it is only relatively recently that its results have left the research laboratories and found their wa ..."
Abstract
 Add to MetaCart
Introduction Machine learning (ML) is devoted to the study and computer modelling of learning processes in their multiple manifestations. Although ML research started with the advent of computers, it is only relatively recently that its results have left the research laboratories and found their way into realworld applications. The motivation for applying ML techniques to realworld tasks is strong: the problems of manually engineering a knowledge base are now well known, and ML offers a technology for assisting in this task; there is vast potential for automatically discovering new knowledge in the recent explosion of available online databases, too large for humans to manually sift through; and the ability of computers to automatically adapt to changing expertise would offer huge benefits for the maintenance and evolution of expert systems. Despite this, the success of ML applications varies tremendously. There are some spectacular successes to its credit, but also the nu