Publications
281 results found
Muggleton SH, Page D, 1995, A Learnability Model for Universal Representations and Its Application to Top-down Induction of Decision Trees., Publisher: Oxford University Press, Pages: 248-267
STERNBERG MJE, MUGGLETON S, SHRINIVASAN A, 1994, DRUG DESIGN BY MACHINE LEARNING, ABSTRACTS OF PAPERS OF THE AMERICAN CHEMICAL SOCIETY, Vol: 208, Pages: 138-COMP, ISSN: 0065-7727
BOOBIS AR, 1994, 3RD INTERNATIONAL-CONFERENCE ON PRACTICAL IN-VITRO TOXICOLOGY - 25-29 JULY 1993, EAST-MIDLANDS-CONFERENCE-CENTER, NOTTINGHAM, UK - PREFACE, TOXICOLOGY IN VITRO, Vol: 8, Pages: 505-505, ISSN: 0887-2333
- Author Web Link
- Cite
- Citations: 2
Muggleton S, 1994, Bayesian inductive logic programming, Pages: 3-11
Inductive Logic Progrdng (ILP) involves the construction of first-order definite clause theories from examples and background knowledge, Unlike both traditional Machine Learning and Computational Learning Theory, ILP is based on lockstep development of Theory, Implementations and Applications. ILP systems have successful applications in the learning of structure-activity rules for drug design, semantic grammars rules, finite element mesh design rules and rules for prediction of protein structure and mutagenic molecules. The strong applications in ILP can be contrasted with relatively weak PAC-learning results (even highlyrestricted forms of logic programs are known to be prediction-hard). It has been recently argued that the mismatch is due to distributional assumptions made in application domains. These assumptions can be modelled as a Bayesiau prior probability representing subjective degrees of belief. Other authors have argued for the use of Bayesian prior distributions for reasons different to those here, though this has not lead to a new model of polynomial-time learnability. Incorporation of Bayesian prior distributions over time-bounded hypotheses in PAC leads to a new model called ULearnability. It is argued that U-learnability is more appropriate than PAC for Universal (Turing computable) languages. Time-bounded logic programs have been shown to be polynomially U-learnable under certain distributions. The use of time-bounded hypotheses enforces decidability and allows a unified characterisation of speed-up learning and inductive learning. U-learnability has as special cases PAC and Natarajan's model of speed-up learning.
STERNBERG MJE, KING RD, LEWIS RA, et al., 1994, APPLICATION OF MACHINE LEARNING TO STRUCTURAL MOLECULAR-BIOLOGY, PHILOSOPHICAL TRANSACTIONS OF THE ROYAL SOCIETY OF LONDON SERIES B-BIOLOGICAL SCIENCES, Vol: 344, Pages: 365-371, ISSN: 0962-8436
- Author Web Link
- Cite
- Citations: 27
Sternberg MJE, Hirst JD, Lewis RA, et al., 1994, Application of machine learning to protein structure prediction and drug design, ISSN: 0963-3308
The use of inductive-based logic programming (ILP) in predicting protein structure and drug design was discussed in this article. In the study of alpha and beta fields with alternating alpha-helices and beta-sheet strands, ILP program GOLEM was employed. GOLEM was capable of generating an inductive hypothesis based on coded facts and negative counter-example and chemical background knowledge. However, for drug design machine learning program PROLOG was applied. PROLOG can comprehend the properties of bonds and atoms. It can also give insight in the chemical principles of aromatic and heteroaromatic nitro compounds. In conclusion, machine learning ILP programs can greatly enhance biochemical developments.
Muggleton S, 1994, Bayesian Inductive Logic Programming, Pages: 371-379
Inductive Logic Programming (ILP) involves the construction of first-order definite clause theories from examples and background knowledge. Unlike both traditional Machine Learning and Computational Learning Theory, ILP is based on lockstep development of Theory, Implementations and Applications. ILP systems have successful applications in the learning of structure-activity rules for drug design, semantic grammars rules, finite element mesh design rules and rules for prediction of protein structure and mutagenic molecules. The strong applications in ILP can be contrasted with relatively weak PAC-learning results (even highly-restricted forms of logic programs are known to be prediction-hard). It has been recently argued that the mismatch is due to distributional assumptions made in application domains. These assumptions can be modelled as a Bayesian prior probability representing subjective degrees of belief. Other authors have argued for the use of Bayesian prior distributions for reasons different to those here, though this has not lead to a new model of polynomial-time learnability. Incorporation of Bayesian prior distributions over time-bounded hypotheses in PAC leads to a new model called U-learnability. It is argued that U-learnability is more appropriate than PAC for Universal (Turing computable) languages. Time-bounded logic programs have been shown to be polynomially U-learnable under certain distributions. The use of time-bounded hypotheses enforces decidability and allows a unified characterisation of speed-up learning and inductive learning. U-learnability has as special cases PAC and Natarajan's model of speed-up learning.
Muggleton S, de Raedt L, 1994, Inductive Logic Programming: Theory and methods, The Journal of Logic Programming, Vol: 19-20, Pages: 629-679, ISSN: 0743-1066
Inductive Logic Programming (ILP) is a new discipline which investigates the inductive construction of first-order clausal theories from examples and background knowledge. We survey the most important theories and methods of this new field. First, various problem specifications of ILP are formalized in semantic settings for ILP, yielding a "model-theory" for ILP. Second, a generic ILP algorithm is presented. Third, the inference rules and corresponding operators used in ILP are presented, resulting in a "proof-theory" for ILP. Fourth, since inductive inference does not produce statements which are assured to follow from what is given, inductive inferences require an alternative form of justification. This can take the form of either probabilistic support or logical constraints on the hypothesis language. Information compression techniques used within ILP are presented within a unifying Bayesian approach to confirmation and corroboration of hypotheses. Also, different ways to constrain the hypothesis language or specify the declarative bias are presented. Fifth, some advanced topics in ILP are addressed. These include aspects of computational learning theory as applied to ILP, and the issue of predicate invention. Finally, we survey some applications and implementations of ILP. ILP applications fall under two different categories: first, scientific discovery and knowledge acquisition, and second, programming assistants. © 1994.
Muggleton S, 1994, Predicate invention and utilization, Journal of Experimental and Theoretical Artificial Intelligence, Vol: 6, Pages: 121-130, ISSN: 0952-813X
Inductive logic programming (ILP) involves the synthesis of logic programs from examples. In terms of scientific theory formation ILP systems define observational predicates in terms of a set of theoretical predicates. However, certain basic theorems indicate that with an inadequate theoretical vocabulary this is not always possible. Predicate invention is the augmentation of a given theoretical vocabulary to allow finite axiomatization of the observational predicates. New theoretical predicates need to be chosen from a well-defined universe of such predicates. In this paper a partial order of utilization is described over such a universe. This ordering is a special case of a logical translation. The notion of utilization allows the definition of an equivalence relationship over new predicates. In a manner analogous to Plotkin, clause refinement is defined relative to given background knowledge and a universe of new predicates. It is shown that relative least clause refinement is defined and unique whenever there exists a relative least general generalization of a set of clauses. Results of a preliminary implementation of this approach are given. © 1994 Taylor & Francis Group, LLC.
Muggleton S, 1994, Inductive logic programming, ACM SIGART Bulletin, Vol: 5, Pages: 5-11, ISSN: 0163-5719
<jats:p> Inductive Logic Programming (ILP) is a research area which investigates the construction of first-order definite clause theories from examples and background knowledge. ILP systems have been applied successfully in a number of real-world domains. These include the learning of structure-activity rules for drug design, finite-element mesh design rules, rules for primary-secondary prediction of protein structure and fault diagnosis rules for satellites. There is a well established tradition of learning-in-the-limit results in ILP. Recently some results within Valiant's PAC-learning framework have also been demonstrated for ILP systems. In this paper it is argued that algorithms can be directly <jats:italic>derived</jats:italic> from the formal specifications of ILP. This provides a common basis for Inverse Resolution, Explanation-Based Learning, Abduction and Relative Least General Generalisation. A new general-purpose, efficient approach to predicate invention is demonstrated. ILP is underconstrained by its logical specification. Therefore a brief overview of extra-logical constraints used in ILP systems is given. Some present limitations and research directions for the field are identified. </jats:p>
MUGGLETON S, KING RD, STERNBERG MJE, 1993, PROTEIN SECONDARY STRUCTURE PREDICTION USING LOGIC-BASED MACHINE LEARNING, PROTEIN ENGINEERING, Vol: 6, Pages: 549-549, ISSN: 0269-2139
- Author Web Link
- Cite
- Citations: 3
Muggleton S, 1993, Inductive logic programming: Derivations, successes and shortcomings, Pages: 21-37, ISSN: 0302-9743
Inductive Logic Programming (ILP) is a research area which investigates the construction of quantified definite clause theories from examples and background knowledge. ILP systems have been applied successfully in a number of real-world domains. These include the learning of structure-activity rules for drug design, finite-element mesh design rules, rules for primary-secondary prediction of protein structure and fault diagnosis rules for satellites. There is a well established tradition of learning-in-the-limit results in ILP. Recently some results within Valiant's PAC-learning frame-work have also been demonstrated for ILP systems. In this paper it is argued that algorithms can be directly derived from the formal specifications of ILP. This provides a common basis for Inverse Resolution, Explanation-Based Learning, Abduction and Relative Least General Generalisation. A new general-purpose, efficient approach to predicate invention is demonstrated. ILP is underconstrained by its logical specification. Therefore a brief overview of extra-logical constraints used in ILP systems is given. Some present limitations and research directions for the field are identified.
Džeroski S, Muggleton S, Russell S, 1993, Learnability of constrained logic programs, Pages: 342-347, ISSN: 0302-9743
The field of Inductive Logic Programming (ILP) is concerned with inducing logic programs from examples in the presence of back-ground knowledge. This paper defines the ILP problem and describes several syntactic restrictions that are often used in ILP. We then derive some positive results concerning the learnability of these restricted classes of logic programs, by reduction to a standard propositional learning problem. More specifically, k-literal predicate definitions consisting of constrained, function-free, nonrecursive program clauses are polynomially PAC-learnable under arbitrary distributions.
Muggleton S, 1993, Optimal layered learning: A PAC approach to incremental sampling, Pages: 37-44, ISSN: 0302-9743
It is best to learn a large theory in small pieces. An approach called “layered learning” starts by learning an approximately correct theory. The errors of this approximation are then used to construct a second-order “correcting” theory, which will again be only approximately correct. The process is iterated until some desired level of overall theory accuracy is met. The main advantage of this approach is that the sizes of successive training sets (errors of the hypothesis from the last iteration) are kept low. General lower-bound PAC-learning results are used in this paper to show that optimal layered learning results in the total training set size (t) increasing linearly in the number of layers. Meanwhile the total training and test set size (m) increases exponentially and the error (ϵ) decreases exponentially. As a consequence, a model of layered learning which requires that t, rather than m, be a polynomial function of the logarithm of the concept space would make learnable many concept classes which are not learnable in Valiant’s PAC model.
King RD, Srinivasan A, Muggleton S, et al., 1993, Drug design using inductive logic programming, Pages: 646-655, ISSN: 1530-1605
Determining the quantitative structure-activity relationship (QSAR) of a related series of drugs is a central aspect of the drug design process. The machine learning program Golem from the field of inductive logic programming (ILP) applied to QSAR. ILP is the most suitable machine learning technique because it can represent the structural and relational aspects of drugs. A five-step methodology for using machine learning in drug design is presented that consists of identification of the problem, choice of a representation, induction, interpretation of results, and synthesis of new drugs.
Muggleton SH, 1993, Inverting Entailment and Progol., Publisher: Oxford University Press, Pages: 135-190
KING RD, MUGGLETON S, LEWIS RA, et al., 1992, DRUG DESIGN BY MACHINE LEARNING - THE USE OF INDUCTIVE LOGIC PROGRAMMING TO MODEL THE STRUCTURE-ACTIVITY-RELATIONSHIPS OF TRIMETHOPRIM ANALOGS BINDING TO DIHYDROFOLATE-REDUCTASE, PROCEEDINGS OF THE NATIONAL ACADEMY OF SCIENCES OF THE UNITED STATES OF AMERICA, Vol: 89, Pages: 11322-11326, ISSN: 0027-8424
- Author Web Link
- Cite
- Citations: 134
MUGGLETON S, KING RD, STERNBERG MJE, 1992, PROTEIN SECONDARY STRUCTURE PREDICTION USING LOGIC-BASED MACHINE LEARNING, PROTEIN ENGINEERING, Vol: 5, Pages: 647-657, ISSN: 0269-2139
- Author Web Link
- Cite
- Citations: 153
Feng C, Muggleton S, 1992, Towards Inductive Generalisation in Higher Order Logic, Pages: 154-162
In many cases, higher order (Horn) clauses are more suited to express certain concepts when relations between predicates exist. However, to date there has been no appropriate higher order formalism within which efficient inductive generalisation can be carried out. This paper describes inductive generalisation in M\ - a higher order formalism which not only retains the expressiveness of -calculus but also provides for effective and efficient inductive generalisation. The main strength of R\ is twofold: it is a higher formalism extension of the (clausal) first order logic and it can be mechanised in a way similar to the first order case in Horn clause form. For a class of restricted M\, their least general generalisation (LGG) is unique, and so is their most general unification. Inductive generalisation in M\ is implemented in the algorithm HOLGG. This algorithm has been applied to some interesting induction problems in the induction of higher order rule templates and automatic program transformation.
Dzeroski S, Muggleton S, Russell S, 1992, PAC-learnability of determinate logic programs, Pages: 128-135
The field of Inductive Logic Programming (ILP) is concerned with inducing logic programs from examples in the presence of background knowledge. This paper defines the ILP problem, and describes the various syntactic restrictions that are commonly used for learning first-order representations. We then derive some positive results concerning the learnability of these restricted classes of logic programs, by reduction to a standard propositional learning problem. More specifically, k-clause predicate definitions consisting of determinate, function-free, non-recursive Horn clauses with variables of bounded depth are polynomially learnable under simple distributions. Similarly, recursive k-clause definitions are polynomially learnable under simple distributions if we allow existential and membership queries about the target concept.
Muggleton S, Srinivasan A, Bain M, 1992, Compression, Significance and Accuracy, Pages: 338-347
Inductive Logic Programming (ILP) involves learning relational concepts from examples and background knowledge. To date all ILP learning systems make use of tests inherited from propositional and decision tree learning for evaluating the significance of hypotheses. None of these significance tests take account of the relevance or utility of the background knowledge. In this paper we describe a method, called HP-compression, of evaluating the significance of a hypothesis based on the degree to which it allows compression of the observed data with respect to the background knowledge. This can be measured by comparing the lengths of the input and output tapes of a reference Turing machine which will generate the examples from the hypothesis and a set of derivational proofs. The model extends an earlier approach of Muggleton by allowing for noise. The truth values of noisy instances are switched by making use of correction codes. The utility of compression as a significance measure is evaluated empirically in three independent domains. In particular, the results show that the existence of positive compression distinguishes a larger number of significant clauses than other significance tests The method is also shown to reliably distinguish artificially introduced noise as incompressible data.
Bain M, Muggleton SH, 1992, Learning optimal chess strategies., Publisher: Clarendon Press, Oxford, Pages: 291-309
STERNBERG MJE, LEWIS RA, KING RD, et al., 1992, MODELING THE STRUCTURE AND FUNCTION OF ENZYMES BY MACHINE LEARNING, FARADAY DISCUSSIONS, Vol: 93, Pages: 269-280, ISSN: 1359-6640
- Author Web Link
- Cite
- Citations: 2
Sternberg MJE, Lewis RA, King RD, et al., 1992, Machine Learning and biomolecular modelling., Publisher: Clarendon Press, Oxford, Pages: 193-212
Srinivasan A, Muggleton SH, Bain M, 1992, The Justification of Logical Theories based on Data Compression., Publisher: Clarendon Press, Oxford, Pages: 87-121
Muggleton SH, 1992, Logic and Learning: Turing's legacy., Publisher: Clarendon Press, Oxford, Pages: 37-56
Muggleton SH, 1992, Developments in Inductive Logic Programming, Panel Position Paper., Publisher: IOS Press, Pages: 1071-1073
Muggleton S, 1991, Inductive logic programming, New Generation Computing, Vol: 8, Pages: 295-318, ISSN: 0288-3635
A new research area, Inductive Logic Programming, is presently emerging. While inheriting various positive characteristics of the parent subjects of Logic Programming and Machine Learning, it is hoped that the new area will overcome many of the limitations of its forebears. The background to present developments within this area is discussed and various goals and aspirations for the increasing body of researchers are identified. Inductive Logic Programming needs to be based on sound principles from both Logic and Statistics. On the side of statistical justification of hypotheses we discuss the possible relationship between Algorithmic Complexity theory and Probably-Approximately-Correct (PAC) Learning. In terms of logic we provide a unifying framework for Muggleton and Buntine's Inverse Resolution (IR) and Plotkin's Relative Least General Generalisation (RLGG) by rederiving RLGG in terms of IR. This leads to a discussion of the feasibility of extending the RLGG framework to allow for the invention of new predicates, previously discussed only within the context of IR. © 1991 Ohmsha, Ltd. and Springer.
Bratko I, Muggleton S, Varšek A, 1991, Learning Qualitative Models of Dynamic Systems, Pages: 385-388
A technique is described for learning qualitative models of dynamic systems. The QSIM formalism is used as a representation for learned qualitative models. The problem of learning QSIM-type models is formulated in logic, and the GOLEM learning program is used for induction. An experiment in learning a qualitative model of the connected containers system, also called U-tube, is described in detail.
Brazdil P, Muggleton S, 1991, Learning to relate terms in a multiple agent environment, Pages: 424-439, ISSN: 0302-9743
In the first part of the paper we describe how different agents can arrive at different (but overlapping) views of reality. Although the agents can cooperate when answering queries, it is often desirable to construct an integrated theory that explains ‘best’ a given reality. The technique of knowledge integration based on an earlier work is briefly reviewed and some shortcomings of this technique are pointed out. One of the assumptions underlying the earlier work was that all agents must use the same predicate vocabulary. Here we are concerned with the problems that can arise if this assumption does not hold. We also show how these problems can be overcome. It is shown that standard machine learning techniques can be used to acquire the meaning of other agent’s concepts. The experiments described in this paper employ INTEG.3, a knowledge integration system, and GOLEM, an inductive system based on relative least general generalization.
This data is extracted from the Web of Science and reproduced under a licence from Thomson Reuters. You may not copy or re-distribute this data in whole or in part without the written consent of the Science business of Thomson Reuters.