Number Of Incorrect Predictions Research Articles

The Lovász Hinge: A Novel Convex Surrogate for Submodular Losses.

Learning with non-modular losses is an important problem when sets of predictions are made simultaneously. The main tools for constructing convex surrogate loss functions for set prediction are margin rescaling and slack rescaling. In this work, we show that these strategies lead to tight convex surrogates iff the underlying loss function is increasing in the number of incorrect predictions. However, gradient or cutting-plane computation for these functions is NP-hard for non-supermodular loss functions. We propose instead a novel surrogate loss function for submodular losses, the Lovász hinge, which leads to O(p logp) complexity with O(p) oracle accesses to the loss function to compute a gradient or cutting-plane. We prove that the Lovász hinge is convex and yields an extension. As a result, we have developed the first tractable convex surrogates in the literature for submodular losses. We demonstrate the utility of this novel convex surrogate through several set prediction tasks, including on the PASCAL VOC and Microsoft COCO datasets.

Does masked and unmasked priming reflect Bayesian inference as implemented in the Bayesian Reader?

Norris and Kinoshita (2008) describe a Bayesian theory of masked and unmasked priming designed to explain a complex pattern of word and nonword priming across a range of tasks. Their theory is implemented within the Bayesian Reader model, and the model makes some predictions that are confirmed in a set of experiments. The authors consider alternative accounts of priming and conclude that only their theory can account for the results obtained. However, contrary to the authors' claims, the Bayesian Reader makes a number of incorrect predictions regarding masked and unmasked priming phenomena, whereas alternative theories can accommodate current findings.