Change of Representation and Inductive Bias

Change. of. Representation. and. Inductive. Bias. One of the most important emerging concerns of machine learning researchers is the dependence of their learning programs on the underlying representations, especially on the languages ...

Author: D. Paul Benjamin

Publisher: Springer Science & Business Media

ISBN: 9781461315230

Category: Computers

Page: 356

View: 648

Download →

Change of Representation and Inductive Bias One of the most important emerging concerns of machine learning researchers is the dependence of their learning programs on the underlying representations, especially on the languages used to describe hypotheses. The effectiveness of learning algorithms is very sensitive to this choice of language; choosing too large a language permits too many possible hypotheses for a program to consider, precluding effective learning, but choosing too small a language can prohibit a program from being able to find acceptable hypotheses. This dependence is not just a pitfall, however; it is also an opportunity. The work of Saul Amarel over the past two decades has demonstrated the effectiveness of representational shift as a problem-solving technique. An increasing number of machine learning researchers are building programs that learn to alter their language to improve their effectiveness. At the Fourth Machine Learning Workshop held in June, 1987, at the University of California at Irvine, it became clear that the both the machine learning community and the number of topics it addresses had grown so large that the representation issue could not be discussed in sufficient depth. A number of attendees were particularly interested in the related topics of constructive induction, problem reformulation, representation selection, and multiple levels of abstraction. Rob Holte, Larry Rendell, and I decided to hold a workshop in 1988 to discuss these topics. To keep this workshop small, we decided that participation be by invitation only.
Posted in:

Machine Learning Proceedings 1990

This a very fundamental sort of representation change; such a change alters the very space over which learning occurs, ... has primarily addressed the problem of changing the representation of hypotheses (i.e., shift of inductive bias.) ...

Author: Machine Learning

Publisher: Morgan Kaufmann

ISBN: 9781483298580

Category: Computers

Page: 427

View: 501

Download →

Machine Learning Proceedings 1990
Posted in:

PRICAI 2000 Topics in Artificial Intelligence

The representation will have all repeated structure separated into other trees, but this is only a change in representation, not concept. We can change the inductive bias of the algorithm, and hence the concept learned, by extending the ...

Author: Riichiro Mizoguchi

Publisher: Springer

ISBN: 9783540445333

Category: Computers

Page: 838

View: 302

Download →

PRICAI 2000, held in Melbourne, Australia, is the sixth Pacific Rim Interna tional Conference on Artificial Intelligence and is the successor to the five earlier PRICAIs held in Nagoya (Japan), Seoul (Korea), Beijing (China), Cairns (Aus tralia) and Singapore in the years 1990, 1992, 1994, 1996 and 1998 respectively. PRICAI is the leading conference in the Pacific Rim region for the presenta tion of research in Artificial Intelligence, including its applications to problems of social and economic importance. The objectives of PRICAI are: To provide a forum for the introduction and discussion of new research results, concepts and technologies; To provide practising engineers with exposure to and an evaluation of evolving research, tools and practices; To provide the research community with exposure to the problems of practical applications of AI; and To encourage the exchange of AI technologies and experience within the Pacific Rim countries. PRICAI 2000 is a memorial event in the sense that it is the last one in the 20"" century. It reflects what researchers in this region believe to be promising for their future AI research activities. In fact, some salient features can be seen in the papers accepted. We have 12 papers on agents, while PRICAI 96 and 98 had no more than two or three. This suggests to us one of the directions in which AI research is going in the next century. It is true that agent research provides us with a wide range of research subjects from basic ones to applications.
Posted in:

Machine Learning Proceedings 1989

Induction over explanations: A method that exploits domain knowledge to learn from examples (Tech. Rep. No. 88-30-3). ... Proceedings of the First International Workshop in Change of Representation and Inductive Bias (pp. 293–305).

Author: Machine Learning

Publisher: Morgan Kaufmann

ISBN: 9781483297408

Category: Computers

Page: 510

View: 415

Download →

Machine Learning Proceedings 1989
Posted in:

Machine Learning ECML 93

In D. Paul Benjamin , editor , Change of Representation and Inductive Bias , pages 267-308 . Kluwer , Boston , 1990 . 21. Cullen Schaffer . When does overfitting decrease prediction accuracy in induced decision trees and rule sets ?

Author: Pavel B. Brazdil

Publisher: Springer Science & Business Media

ISBN: 3540566023

Category: Computers

Page: 469

View: 846

Download →

This volume contains the proceedings of the Eurpoean Conference on Machine Learning (ECML-93), continuing the tradition of the five earlier EWSLs (European Working Sessions on Learning). The aim of these conferences is to provide a platform for presenting the latest results in the area of machine learning. The ECML-93 programme included invited talks, selected papers, and the presentation of ongoing work in poster sessions. The programme was completed by several workshops on specific topics. The volume contains papers related to all these activities. The first chapter of the proceedings contains two invited papers, one by Ross Quinlan and one by Stephen Muggleton on inductive logic programming. The second chapter contains 18 scientific papers accepted for the main sessions of the conference. The third chapter contains 18 shorter position papers. The final chapter includes three overview papers related to the ECML-93 workshops.
Posted in:

A Compendium of Machine Learning

In D. Paul Benjamin , editor , Change in Representation and Inductive Bias , pp . 247-266 . Kluwer Academic Publishers , Boston , 1990 . [ 363 ] A. Giordana , D. Roverso , and L. Saitta . Abstracting Background Knowledge for Concept ...

Author: Garry Briscoe

Publisher: Intellect Books

ISBN: 1567501796

Category: Computers

Page: 353

View: 855

Download →

Machine learning is a relatively new branch of artificial intelligence. The field has undergone a significant period of growth in the 1990s, with many new areas of research and development being explored.
Posted in:

Multistrategy Learning

Declarative bias: An overview. In P. Benjamin (Ed.), Change of representation and inductive bias. Norwell, MA: Kluwer Academic Press. Schank, R., & Albelson, R. (1977). Scripts, plans, goals, and understanding.

Author: Ryszard S. Michalski

Publisher: Springer Science & Business Media

ISBN: 9781461532026

Category: Computers

Page: 155

View: 432

Download →

Most machine learning research has been concerned with the development of systems that implememnt one type of inference within a single representational paradigm. Such systems, which can be called monostrategy learning systems, include those for empirical induction of decision trees or rules, explanation-based generalization, neural net learning from examples, genetic algorithm-based learning, and others. Monostrategy learning systems can be very effective and useful if learning problems to which they are applied are sufficiently narrowly defined. Many real-world applications, however, pose learning problems that go beyond the capability of monostrategy learning methods. In view of this, recent years have witnessed a growing interest in developing multistrategy systems, which integrate two or more inference types and/or paradigms within one learning system. Such multistrategy systems take advantage of the complementarity of different inference types or representational mechanisms. Therefore, they have a potential to be more versatile and more powerful than monostrategy systems. On the other hand, due to their greater complexity, their development is significantly more difficult and represents a new great challenge to the machine learning community. Multistrategy Learning contains contributions characteristic of the current research in this area.
Posted in:

Topics in Artificial Intelligence

In D. P. Benjamin ( Eds . ) , Change of Representation and Inductive Bias ( pp . 147-167 ) . Boston : Kluwer Academic Publishers . 27. Winston , M. E. , Chaffin , R. , and Hermann , D. J. ( 1987 ) . A taxonomy of PartWhole relations .

Author: Italian Association for Artificial Intelligence. Congress

Publisher: Springer Science & Business Media

ISBN: 3540604375

Category: Computers

Page: 450

View: 484

Download →

This book presents the refereed proceedings of the 4th Congress of the Italian Association for Artificial Intelligence, AI*IA '95, held in Florence, Italy, in October 1995. The 31 revised full papers and the 12 short presentations contained in the volume were selected from a total of 101 submissions on the basis of a careful reviewing process. The papers are organized in sections on natural language processing, fuzzy systems, machine learning, knowledge representation, automated reasoning, cognitive models, robotics and planning, connectionist models, model-based reasoning, and distributed artificial intelligence.
Posted in:

Inductive Logic Programming

In Proceedings of the First International Workshop in Change of Representation and Inductive Bias , page 279 , Phillips Laboratories , Briarcliffe Manor , June 1988 . [ 10 ] D. Osherson and S. Weinstein . Paradigms of truth detection ...

Author: Stephen Muggleton

Publisher: Morgan Kaufmann

ISBN: 0125097158

Category: Computers

Page: 565

View: 162

Download →

Inductive logic programming is a new research area formed at the intersection of machine learning and logic programming. While the influence of logic programming has encouraged the development of strong theoretical foundations, this new area is inheriting its experimental orientation from machine learning. Inductive Logic Programming will be an invaluable text for all students of computer science, machine learning and logic programming at an advanced level. * * Examination of the background to current developments within the area * Identification of the various goals and aspirations for the increasing body of researchers in inductive logic programming * Coverage of induction of first order theories, the application of inductive logic programming and discussion of several logic learning programs * Discussion of the applications of inductive logic programming to qualitative modelling, planning and finite element mesh design
Posted in: