The combining classifier: to train or not to train?

By Robert P. W. Duin. Object recognition supported by user interaction for service robots, 2002

412 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

An Off-line Signature Verification System Based on Fusion of Local and Global Information (2004, background): “Both approaches have their advantages and disadvantages as discussed in [25].”
Getting the Most Out of Ensemble Selection (2006, result): “Counter to our and others’ intuition [5], calibrating models to put all predictions on the same scale before averaging them did not improve ensemble selection’s effectiveness.”
Multi-column deep neural network for traffic sign classification (2012, background): “A key question is whether to optimize the combination of outputs of various models or not (Duin, 2002).”
A survey of multiple classifier systems as hybrid systems (2014, background): “little practical applicability because of the hard conditions imposed by them [145].”
Cost-sensitive decision tree ensembles for effective imbalanced classification (2014, positive): “While simple aggregation methods (like minimum, maximum, product, mean) can be used, they are typically subject to rather restrictive conditions [12] which limit their practical use.”