No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
A Hill-Climbing Landmarker Generation Algorithm Based on Efficiency and Correlativity Criteria.
Abstract
For a given classification task, there are typically several learning algorithms available. The question then arises: which is the most appropriate algorithm to apply. Recently, we proposed a new algorithm for making such a selection based on landmarking - a meta-learning strategy that utilises meta-features that are measurements based on efficient learning algorithms. This algorithm, which creates a set of landmarkers that each utilise subsets of the algorithms being landmarked, was shown to be able to estimate accuracy well, even when employing a small fraction of the given algorithms. However, that version of the algorithm has exponential computational complexity for training. In this paper, we propose a hill-climbing version of the landmarker generation algorithm, which requires only polynomial training time complexity. Our experiments show that the landmarkers formed have similar results to the more complex version of the algorithm.
... The actual generation method is discussed in greater detail in (Ler et al., 2004a, 2004b), while a more efficient hill-climbing version of the proposed landmarker generation algorithm is described in (Ler et al., 2005). ...
... The actual generation method is discussed in greater detail in (Ler et al., 2004aLer et al., , 2004b), while a more efficient hill-climbing version of the proposed landmarker generation algorithm is described in (Ler et al., 2005). ...
Recent advances in meta-learning are providing the foundations to construct meta-learning assistants and task-adaptive learners. The goal of this special issue is to foster an interest in meta-learning by compiling representative work in the field. The contributions to this special issue provide strong insights into the construction of future meta-learning tools. In this introduction we present a common frame of reference to address work in meta-learning through the concept of meta-knowledge. We show how meta-learning can be simply defined as the process of exploiting knowledge about learning that enables us to understand and improve the performance of learning algorithms.
Machine-learning research has been making great progress in many directions. This article summarizes four of these directions and discusses some current open problems. The four directions are (1) the improvement of classification accuracy by learning ensembles of classifiers, (2) methods for scaling up supervised learning algorithms, (3) reinforcement learning, and (4) the learning of complex stochastic models.
The selection of an appropriate inducer is crucial for performing effective classification. In previous work we presented a system called NOEMON which relied on a mapping between dataset characteristics and inducer performance to propose inducers for specific datasets. Instance-based learning was used to create that mapping. Here we explore the use of decision trees inducers as the inducers on the meta-learning level. We believe that they posses a set of properties that match the properties of the meta-learning problem that we are trying to solve. The results show that the performance of the system is indeed improved with the use of the decision tree learners.
The selection of an appropriate inducer is crucial for performing
effective classification. In previous work we presented a system called
NOEMON which relied on a mapping between dataset characteristics and
inducer performance to propose inducers for specific datasets. Instance
based learning was used to create that mapping. Here we extend and
refine the set of data characteristics; we also use a wider range of
base-level inducers and a much larger collection of datasets to create
the meta-models. We compare the performance of meta-models produced by
instance based learners, decision trees and boosted decision trees. The
results show that decision trees and boosted decision trees models
enhance the perfomance of the system