Logistic Disjunctive Normal Network (LDNN) is a novel classification scheme, which consists of one adaptive layer of feature detectors implemented by logistic sigmoid functions followed by two fixed layers of logical units that compute conjunctions and disjunctions, respectively. LDNN is a general binary classifier that is designed to find the classification function in disjunctive normal form. In short, LDNN forms the decision function by disjunction of conjunction of half-spaces defined by perceptrons and approximated by logistic neurons. The training is done by back-propagation of the error in order to minimize an objective function. LDNN achieves state-of-the-art results on general classification problems. It is a flexible and promising structure that can be used in a variety of structures for different machine learning tasks including regression, object recognition, scene labeling and shape modeling.

Disjunctive Normal Shape Model (DNSM) is a differentiable implicit and parametric shape model where the parameters of the model are the angles and positions of linear discriminants which form the boundary of the shape.