We present a new method for speaker verification that uses the diversity of information from multiple feature representations. The principle behind the method is that certain feat...
Abstract. We consider the problem of learning an unknown (overcomplete) basis from an unknown sparse linear combination. Introducing the "sparse coding neural gas" algori...
Discovering a representation that allows auditory data to be parsimoniously represented is useful for many machine learning and signal processing tasks. Such a representation can b...
– With the World Wide Web popularity the information retrieval area has a new challenge intending to retrieve information resources by their meaning by using a knowledge base. No...
We argue that producing maintainable high-performance implementations of finite element methods for multiple targets requires that they are written using a high-level domain-speci...