Computational Learning Theory and Natural Learning Systems: Volume 2 Intersections between Theory and Experiment(A Bradford Book)

As with Volume I, this second volume represents a synthesis of issues in three historically distinct areas of learning research: computational learning theory, neural network research, and symbolic machine learning. While the first volume provided a forum for building a science of computational learning across fields, this volume attempts to define plausible areas of joint research: the contributions are concerned with finding constraints for theory while at the same time interpreting theoretic results in the context of experiments with actual learning systems. Subsequent volumes will focus on areas identified as research opportunities.Computational learning theory, neural networks, and AI machine learning appear to be disparate fields; in fact they have the same goal: to build a machine or program that can learn from its environment. Accordingly, many of the papers in this volume deal with the problem of learning from examples. In particular, they are intended to encourage discussion between those trying to build learning algorithms (for instance, algorithms addressed by learning theoretic analyses are quite different from those used by neural network or machine-learning researchers) and those trying to analyze them.The first section provides theoretical explanations for the learning systems addressed, the second section focuses on issues in model selection and inductive bias, the third section presents new learning algorithms, the fourth section explores the dynamics of learning in feedforward neural networks, and the final section focuses on the application of learning algorithms.A Bradford Book

Table of Contents:
Part 1 Learning theory: Bayes decisions in a neural network-PAC setting, Svetlana Anulova et al; average case analysis of kappa-CNF and kappa-DNF learning algorithms, Daniel S. Hirschberg et al; filter likelihoods and exhaustive learning, David H. Wolpert. Part 2 Model selection and inductive bias: incorporating prior knowledge into networks of locally-tuned units, Martin Roscheisen et al; using knowledge-based neural networks to refine roughly-correct information, Geoffrey G. Towell and Jude W. Shavlik; sensitivity constraints in learning, Scott H. Clearwater and Yongwon Lee; evaluation of learning biases using probabilistic domain knowledge, Marie desJardins; detecting structure in small datasets by network fitting under complexity constraints, W. Finnoff and H.G. Zimmermann; associative methods in reinforcement learning - an empirical study, Leslie Pack Kaelbling. Part 3 Learning algorithms: a schema for using multiple knowledge, Matjaz Gams et al; probabilistic hill-climbing, William W. Cohen et al; prototype selection using competitive learning, Michael Lemmon; learning with instance-based encodings, Henry Tirri; contrastive learning with graded random networks, Javier R. Movellan and James L. McClelland; probability density estimation and local basis function neural networks, Padhraic Smyth. Part 4 Dynamics of learning: Hamiltonian dynamics of neural networks, Ulrich Ramacher; learning properties of multi-layer perceptrons with and without feedback, D. Gawronska et al. Part 5 Applications: unsupervised learning for mobile robot navigation using probabilistic data association, Ingemar J. Cox and John J. Leonard; evolution of a subsumption architecture that performs a wall following task for an autonomous mobile robot, John R. Koza; a connectionist model of the learning of personal pronouns in English, Thomas R. Shultz et al; neural network modelling of physiological processes, Volker Tresp et al; projection pursuit learning - some theoretical issues, Ying Zhao and Christopher G. Atkeson; a comparative study of the Kohonen self-organizing map and the elastic net, Yiu-fai Wong.

About the Author :
Stephen José Hanson is Professor of Psychology (Newark Campus) and Member of the Cognitive Science Center (New Brunswick Campus) at Rutgers University. Michael J. Kearns is Professor of Computer and Information Science at the University of Pennsylvania. Ronald L. Rivest is Andrew and Erna Viterbi Professor of Electrical Engineering and Computer Science at the Massachusetts Institute of Technology.