Algorithmic Learning Theory: 16th International Conference, ALT 2005, Singapore, October 8-11, 2005. Proceedings

Sanjay Jain, Hans Ulrich Simon, Etsuji Tomita (auth.), Sanjay Jain, Hans Ulrich Simon, Etsuji Tomita (eds.)354029242X, 9783540292425

This book constitutes the refereed proceedings of the 16th International Conference on Algorithmic Learning Theory, ALT 2005, held in Singapore in October 2005.

The 30 revised full papers presented together with 5 invited papers and an introduction by the editors were carefully reviewed and selected from 98 submissions. The papers are organized in topical sections on kernel-based learning, bayesian and statistical models, PAC-learning, query-learning, inductive inference, language learning, learning and logic, learning from expert advice, online learning, defensive forecasting, and teaching.

---

Table of contents :
Front Matter….Pages –
Editors’ Introduction….Pages 1-9
Invention and Artificial Intelligence….Pages 10-10
The Arrowsmith Project: 2005 Status Report….Pages 11-11
The Robot Scientist Project….Pages 12-12
Algorithms and Software for Collaborative Discovery from Autonomous, Semantically Heterogeneous, Distributed Information Sources….Pages 13-44
Training Support Vector Machines via SMO-Type Decomposition Methods….Pages 45-62
Learning Attribute-Efficiently with Corrupt Oracles….Pages 183-197
Learning DNF by Statistical and Proper Distance Queries Under the Uniform Distribution….Pages 198-210
Learning of Elementary Formal Systems with Two Clauses Using Queries….Pages 211-225
Gold-Style and Query Learning Under Various Constraints on the Target Class….Pages 226-240
Non U-Shaped Vacillatory and Team Learning….Pages 241-255
Measuring Statistical Dependence with Hilbert-Schmidt Norms….Pages 63-77
An Analysis of the Anti-learning Phenomenon for the Class Symmetric Polyhedron….Pages 78-91
Learning Causal Structures Based on Markov Equivalence Class….Pages 92-106
Stochastic Complexity for Mixture of Exponential Families in Variational Bayes….Pages 107-121
ACME: An Associative Classifier Based on Maximum Entropy Principle….Pages 122-134
Constructing Multiclass Learners from Binary Learners: A Simple Black-Box Analysis of the Generalization Errors….Pages 135-147
On Computability of Pattern Recognition Problems….Pages 148-156
PAC-Learnability of Probabilistic Deterministic Finite State Automata in Terms of Variation Distance….Pages 157-170
Learnability of Probabilistic Automata via Oracles….Pages 171-182
Online Allocation with Risk Information….Pages 343-355
Defensive Universal Learning with Experts….Pages 356-370
On Following the Perturbed Leader in the Bandit Setting….Pages 371-385
Mixture of Vector Experts….Pages 386-398
Teaching Learners with Restricted Mind Changes….Pages 474-489
On-line Learning with Delayed Label Feedback….Pages 399-413
Monotone Conditional Complexity Bounds on Future Prediction Errors….Pages 414-428
Learning Multiple Languages in Groups….Pages 256-268
Non-asymptotic Calibration and Resolution….Pages 429-443
Defensive Prediction with Expert Advice….Pages 444-458
Defensive Forecasting for Linear Protocols….Pages 459-473
Inferring Unions of the Pattern Languages by the Most Fitting Covers….Pages 269-282
Identification in the Limit of Substitutable Context-Free Languages….Pages 283-296
Algorithms for Learning Regular Expressions….Pages 297-311
A Class of Prolog Programs with Non-linear Outputs Inferable from Positive Data….Pages 312-326
Absolute Versus Probabilistic Classification in a Logical Setting….Pages 327-342
Back Matter….Pages –