Preserving activations in recurrent neural networks based on surprisal
Sign Up to like & getrecommendations! Published in 2019 at "Neurocomputing"
DOI: 10.1016/j.neucom.2018.11.092
Abstract: Abstract Learning hierarchical abstractions from sequences is a challenging and open problem for recurrent neural networks (RNNs). This is mainly due to the difficulty of detecting features that span over long time distances with also… read more here.
Keywords: activations recurrent; neural networks; preserving activations; recurrent neural ... See more keywords