Algorithmic Learning Theory

Invited Lectures.- Tailoring Representations to Different Requirements.- Theoretical Views of Boosting and Applications.- Extended Stochastic Complexity and Minimax Relative Loss Analysis.- Regular Contributions.- Algebraic Analysis for Singular Statistical Estimation.- Generalization Error of Linear Neural Networks in Unidentifiable Cases.- The Computational Limits to the Cognitive Power of the Neuroidal Tabula Rasa.- The Consistency Dimension and Distribution-Dependent Learning from Queries (Extended Abstract).- The VC-Dimension of Subclasses of Pattern Languages.- On the V ? Dimension for Regression in Reproducing Kernel Hilbert Spaces.- On the Strength of Incremental Learning.- Learning from Random Text.- Inductive Learning with Corroboration.- Flattening and Implication.- Induction of Logic Programs Based on ?-Terms.- Complexity in the Case Against Accuracy: When Building One Function-Free Horn Clause Is as Hard as Any.- A Method of Similarity-Driven Knowledge Revision for Type Specializations.- PAC Learning with Nasty Noise.- Positive and Unlabeled Examples Help Learning.- Learning Real Polynomials with a Turing Machine.- Faster Near-Optimal Reinforcement Learning: Adding Adaptiveness to the E3 Algorithm.- A Note on Support Vector Machine Degeneracy.- Learnability of Enumerable Classes of Recursive Functions from “Typical” Examples.- On the Uniform Learnability of Approximations to Non-recursive Functions.- Learning Minimal Covers of Functional Dependencies with Queries.- Boolean Formulas Are Hard to Learn for Most Gate Bases.- Finding Relevant Variables in PAC Model with Membership Queries.- General Linear Relations among Different Types of Predictive Complexity.- Predicting Nearly as Well as the Best Pruning of a Planar Decision Graph.- On Learning Unionsof Pattern Languages and Tree Patterns.

Includes supplementary material: sn.pub/extras