Active learning strategies can be useful when manual labeling
effort is scarce, as they select the most informative
examples to be annotated first. However, for visual category
...
Sudheendra Vijayanarasimhan (University of Texas a...
In this paper, we propose a sharpness dependent filter design based on the fairing of surface normal, whereby the filtering algorithm automatically selects a filter. This may be a...
Tiling is a widely used loop transformation for exposing/exploiting parallelism and data locality. Effective use of tiling requires selection and tuning of the tile sizes. This is...
In many real world applications, active selection of training examples can significantly reduce the number of labelled training examples to learn a classification function. Differ...
Abstract: We investigate the structure of model selection problems via the bias/variance decomposition. In particular, we characterize the essential structure of a model selection ...