no code implementations • 31 Oct 2016 • Eder Santana, Matthew Emigh, Pablo Zegers, Jose C. Principe
We propose a convolutional recurrent neural network, with Winner-Take-All dropout for high dimensional unsupervised feature learning in multi-dimensional time series.
no code implementations • 22 Mar 2016 • Eder Santana, Matthew Emigh, Jose C. Principe
We propose Information Theoretic-Learning (ITL) divergence measures for variational regularization of neural networks.