Class and subclass probability re-estimation to adapt a classifier in the presence of concept drift

Online Optimization Methods for the Quantification Problem

“…The smoothed versions of p(c) andp(c) are then used in place of the non-smoothed versions in Equation 1. We can show that, as a result, KLD is always bounded by KLD(ps,ps) ≤ O log 1 However, we note that the smoothed KLD still returns a value of 0 when p andp are identical distributions.…”

“…Quantification has independently been studied within statistics [16,22], machine learning [2,8,29], and data mining [10,11]. Unsurprisingly, given this varied literature, quantification also goes under different names, such as counting [23], class probability re-estimation [1], class prior estimation [6], and learning of class balance [8].…”

“…For instance, Balikas et al [2] use quantification for model selection in supervised learning, by tuning hyperparameters that yield the best quantification accuracy on validation data; this allows hyperparameter tuning to be performed without incurring the costs inherent to k-fold cross-validation. Saerens et al [29], followed by other authors [1,32,34], apply quantification to customize a trained classifier to the class prevalence exhibited in the test set, with the goal of improving classification accuracy on unlabeled data exhibiting a class distribution different from that of the training set. The work of Chan and Ng [6] may be seen as a direct application of this notion, as they use quantification to tune a word sense disambiguator to the estimated sense priors of the test set.…”

“…This is not just a feature of KLD, but an inherent feature of any measure for evaluating quantification. In fact, how the error made on a given unlabeled item impacts the overall quantification error depends on how the other items have been classified 1 ; e.g., if FP > FN for the other unlabeled items, then generating an additional false negative is actually beneficial to the overall quantification accuracy, be it measured via KLD or via any other function.…”

A Review on Quantification Learning

“…Methods of both groups are usually not compared; in fact, the performance of the latter group has been normally studied just in terms of the improvement on classification tasks, but not as quantifiers. Unsurpris-ingly given this scenario, algorithms that can be applied for quantification tasks appear on papers that use different keywords and names, such as prior probability shift [Storkey 2009], posterior probability estimation [Alaiz-Rodríguez et al 2011], class prior estimation [Du Plessis and Sugiyama 2014a], class-prior change [Du Plessis and Sugiyama 2014b], prevalence estimation or class ratio estimation [Asoh et al 2012], just to cite some of them.…”