Europe
Separate and Reconstruct: Asymmetric Encoder-Decoder for Speech Separation
In speech separation, time-domain approaches have successfully replaced the time-frequency domain with latent sequence feature from a learnable encoder.
Graph-based Discriminators: Sample Complexity and Expressiveness
Abasic question inlearning theory istoidentify iftwodistributions areidentical when we have access only to examples sampled from the distributions.
Untangling in Invariant Speech Recognition
Cory Stephenson, Jenelle Feather, Suchismita Padhy, Oguz Elibol, Hanlin Tang, Josh McDermott, SueYeon Chung
Meanwhile, deep neural networks have also achieved impressive performance in audio processing applications, both as sub-components of larger systems and as complete end-to-end systems by themselves.
Generalization in Reinforcement Learning with Selective Noise Injection and Information Bottleneck
Maximilian Igl, Kamil Ciosek, Yingzhen Li, Sebastian Tschiatschek, Cheng Zhang, Sam Devlin, Katja Hofmann
Neural Information Processing Systems http://nips.cc/
SLANG: Fast Structured Covariance Approximations for Bayesian Deep Learning with Natural Gradient
Aaron Mishkin, Frederik Kunstner, Didrik Nielsen, Mark Schmidt, Mohammad Emtiyaz Khan
Uncertainty estimation in large deep-learning models is a computationally challenging task, where it is difficult to form even a Gaussian approximation to the posteriordistribution.
Partially-Supervised Image Captioning
Peter Anderson, Stephen Gould, Mark Johnson
Neural Information Processing Systems http://nips.cc/
Characterizing the Exact Behaviors of Temporal Difference Learning Algorithms Using Markov Jump Linear System Theory
Neural Information Processing Systems http://nips.cc/