Bidirectional Recurrent Neural Networks as Generative Models
Annual Conference on Neural Information Processing Systems, 2015.
Bidirectional recurrent neural networks (RNN) are trained to predict both in the positive and negative time directions simultaneously. They have not been used commonly in unsupervised tasks, because a probabilistic interpretation of the model has been difficult. Recently, two different frameworks, GSN and NADE, provide a connection betwee...More
PPT (Upload PPT)