4 papers
Improving Autoencoder Training Performance for Hyperspectral Unmixing with Network Reinitialisation
Kamil Książek, Przemysław Głomb, Michał Romaszewski +3
Neural networks, in particular autoencoders, are one of the most promising solutions for unmixing hyperspectral data, i.e. reconstructing the spectra of observed substances (endmem…
Encouraging an Appropriate Representation Simplifies Training of Neural Networks
Krisztian Buza
A common assumption about neural networks is that they can learn an appropriate internal representations on their own, see e.g. end-to-end learning. In this work we challenge this…
Resting state fMRI functional connectivity-based classification using a convolutional neural network architecture
Regina Meszlényi, Krisztian Buza, Zoltán Vidnyánszky
Machine learning techniques have become increasingly popular in the field of resting state fMRI (functional magnetic resonance imaging) network based classification. However, the a…
Using Echo State Networks for Cryptography
Rajkumar Ramamurthy, Christian Bauckhage, Krisztian Buza +1
Echo state networks are simple recurrent neural networks that are easy to implement and train. Despite their simplicity, they show a form of memory and can predict or regenerate se…