Computational vision often needs to deal with derivatives of digital images. Derivatives are not intrinsic properties of a digital image; a paradigm is required to make them well-deened. Normally, a linear ltering is applied. This can be formulated in terms of scale space, functional minimization or edge detection lters. In this paper, we take regularization (or functional minimization) as a st...

Smoothness regularization is a popular method to decrease generalization error. We propose a novel regularization technique that rewards local distributional smoothness (LDS), a KLdistance based measure of the model’s robustness against perturbation. The LDS is defined in terms of the direction to which the model distribution is most sensitive in the input space. We call the training with LDS r...

NLP models have many and sparse features, and regularization is key for balancing model overfitting versus underfitting. A recently repopularized form of regularization is to generate fake training data by repeatedly adding noise to real data. We reinterpret this noising as an explicit regularizer, and approximate it with a second-order formula that can be used during training without actually ...

To minimize or upper-bound the value of a function “robustly”, we might instead minimize or upper-bound the “ -robust regularization”, defined as the map from a point to the maximum value of the function within an -radius. This regularization may be easy to compute: convex quadratics lead to semidefinite-representable regularizations, for example, and the spectral radius of a matrix leads to ps...

Transductive SVM (TSVM) is a well known semi-supervised large margin learning method for binary text classification. In this paper we extend this method to multi-class and hierarchical classification problems. We point out that the determination of labels of unlabeled examples with fixed classifier weights is a linear programming problem. We devise an efficient technique for solving it. The met...

The Co-Training algorithm uses unlabeled examples in multiple views to bootstrap classifiers in each view, typically in a greedy manner, and operating under assumptions of view-independence and compatibility. In this paper, we propose a Co-Regularization framework where classifiers are learnt in each view through forms of multi-view regularization. We propose algorithms within this framework th...

Information regularization is a principle for assigning labels to unlabeled data points in a semi-supervised setting. The broader principle is based on finding labels that minimize the information induced between examples and labels relative to a topology over the examples; any label variation within a small local region of examples ties together the identities of examples and their labels. Suc...