Machine Learning: ECML 2006

by Furnkranz, Johannes; Scheffer, Tobias; Spiliopoulou, Myra

• ISBN13: 9783540453758
• ISBN10: 354045375X
• Format: Paperback
• Copyright: 2006-11-03
• Publisher: Springer-Verlag New York Inc

Summary

This book constitutes the refereed proceedings of the 17th European Conference on Machine Learning, ECML 2006, held in Berlin, Germany in September 2006, jointly with PKDD 2006.The 46 revised full papers and 36 revised short papers presented together with abstracts of 5 invited talks were carefully reviewed and selected from 564 papers submitted to both, ECML and PKDD. The papers present a wealth of new results in the area and address all current issues in machine learning.

Table of Contents

On temporal evolution in data streams	p. 1
The future of CiteSeer : CiteSeer[superscript x]	p. 2
Learning to have fun	p. 3
Winning the DARPA grand challenge	p. 4
Challenges of urban sensing	p. 5
Learning in one-shot strategic form games	p. 6
A selective sampling strategy for label ranking	p. 18
Combinatorial Markov random fields	p. 30
Learning stochastic tree edit distance	p. 42
Pertinent background knowledge for learning protein grammars	p. 54
Improving Bayesian network structure search with random variable aggregation hierarchies	p. 66
Sequence discrimination using phase-type distributions	p. 78
Languages as hyperplanes : grammatical inference with string kernels	p. 90
Toward robust real-world inference : a new perspective on explanation-based learning	p. 102
Fisher kernels for relational data	p. 114
Evaluating misclassifications in imbalanced data	p. 126
Improving control-knowledge acquisition for planning by active learning	p. 138
PAC-learning of Markov models with hidden state	p. 150
A discriminative approach for the retrieval of images from text queries	p. 162
TildeCRF : conditional random fields for logical sequences	p. 174
Unsupervised multiple-instance learning for functional profiling of genomic data	p. 186
Bayesian learning of Markov network structure	p. 198
Approximate policy iteration for closed-loop learning of visual tasks	p. 210
Task-driven discretization of the joint space of visual percepts and continuous actions	p. 222
EM algorithm for symmetric causal independence models	p. 234
Deconvolutive clustering of Markov states	p. 246
Patching approximate solutions in reinforcement learning	p. 258
Fast variational inference for Gaussian process models through KL-correction	p. 270
Bandit based Monte-Carlo planning	p. 282
Bayesian learning with mixtures of trees	p. 294
Transductive Gaussian process regression with automatic model selection	p. 306
Efficient convolution kernels for dependency and constituent syntactic trees	p. 318
Why is rule learning optimistic and how to correct it	p. 330
Automatically evolving rule induction algorithms	p. 341
Bayesian active learning for sensitivity analysis	p. 353
Mixtures of Kikucki approximations	p. 365
Boosting in PN spaces	p. 377
Prioritizing point-based POMDP solvers	p. 389
Graph based semi-supervised learning with sharper edges	p. 401
Margin-based active learning for structured output spaces	p. 413
Skill acquisition via transfer learning and advice taking	p. 425
Constant rate approximate maximum margin algorithms	p. 437
Batch classification with applications in computer aided diagnosis	p. 449
Improving the ranking performance of decision trees	p. 461
Multiple-instance learning via random walk	p. 473
Localized alternative cluster ensembles for collaborative structuring	p. 485
Distributional features for text categorization	p. 497
Subspace metric ensembles for semi-supervised clustering of high dimensional data	p. 509
An adaptive kernel method for semi-supervised clustering	p. 521
To select or to weigh : a comparative study of model selection and model weighing for SPODE ensembles	p. 533
Ensembles of nearest neighbor forecasts	p. 545
Learning process models with missing data	p. 557
Case-based label ranking	p. 566
Cascade evaluation of clustering algorithms	p. 574
Making good probability estimates for regression	p. 582
Fast spectral clustering of data using sequential matrix compression	p. 590
An information-theoretic framework for high-order co-clustering of heterogeneous objects	p. 598
Efficient inference in large conditional random fields	p. 606
A kernel-based approach to estimating phase shifts between irregularly sampled time series : an application to gravitational lenses	p. 614
Cost-sensitive decision tree learning for forensic classification	p. 622
The minimum volume covering ellipsoid estimation in kernel-defined feature spaces	p. 630
Right of inference : nearest rectangle learning revisited	p. 638
Reinforcement learning for MDPs with constraints	p. 646
Efficient non-linear control through neuroevolution	p. 654
Efficient prediction-based validation for document clustering	p. 663
On testing the missing at random assumption	p. 671
B-matching for spectral clustering	p. 679
Multi-class ensemble-based active learning	p. 687
Active learning with irrelevant examples	p. 695
Classification with support hyperplanes	p. 703
(Agnostic) PAC learning concepts in higher-order logic	p. 711
Evaluating feature selection for SVMs in high dimensions	p. 719
Revisiting Fisher kernels for document similarities	p. 727
Scaling model-based average-reward reinforcement learning for product delivery	p. 735
Robust probabilistic calibration	p. 743
Missing data in kernel PCA	p. 751
Exploiting extremely rare features in text categorization	p. 759
Efficient large scale linear programming support vector machines	p. 767
An efficient approximation to lookahead in relational learners	p. 775
Improvement of systems management policies using hybrid reinforcement learning	p. 783
Diversified SVM ensembles for large data sets	p. 792
Dynamic integration with random forests	p. 801
Bagging using statistical queries	p. 809
Guiding the search in the NO region of the phase transition problem with a partial subsumption test	p. 817
Spline embedding for nonlinear dimensionality reduction	p. 825
Cost-sensitive learning of SVM for ranking	p. 833
Variational Bayesian Dirichlet-multinomial allocation for exponential family mixtures	p. 841
Table of Contents provided by Blackwell. All Rights Reserved.