Latest Generative Adversarial Networks (GANs) are gathering outstanding results through a large-scale training, thus employing models composed of millions parameters requiring extensive computational capabilities. Building such huge undermines their replicability and increases the training instability. Moreover, multi-channel data, as images or audio, usually processed by realvalued convolution...

Learning a disentangled representation is still challenge in the field of interpretability generative adversarial networks (GANs). This paper proposes generic method to modify traditional GAN into an interpretable GAN, which ensures that filters intermediate layer generator encode localized visual concepts. Each filter supposed consistently generate image regions corresponding same concept when...

The ever-increasing size of modern data sets combined with the difficulty of obtaining label information has made semi-supervised learning one of the problems of significant practical importance in modern data analysis. We revisit the approach to semi-supervised learning with generative models and develop new models that allow for effective generalisation from small labelled data sets to large ...

Several frameworks exist to describe how procedural content can be understood, or how it can be used in games. In this paper, we present a framework that considers generativity as a pipeline of successive data transformations, with each transformation either generating, transforming, or pruning away information. This framework has been iterated through repeated engagement and education interact...

This paper describes an incremental parsing approach where parameters are estimated using a variant of the perceptron algorithm. A beam-search algorithm is used during both training and decoding phases of the method. The perceptron approach was implemented with the same feature set as that of an existing generative model (Roark, 2001a), and experimental results show that it gives competitive pe...

We propose a framework for generating samples from probability distribution that differs the of training set. use an adversarial process simultaneously trains three networks, generator and two discriminators. refer to this new model as regularized generative network (RegGAN). evaluate RegGAN on synthetic dataset composed gray scale images we further show it can be used learn some pre-specified ...