Convolutional networks for images, speech, and time series

TL;DR: This survey conducts a systematic survey with comparisons and discussions of various non-autoregressive translation (NAT) models from different aspects, and categorizes the efforts of NAT into several groups, including data manipulation, modeling methods, training criterion, decoding algorithms, and the benefit from pre-trained models.

TL;DR: The system developed at LIF for the SemEval-2016 evaluation campaign to identify sentiment polarity in tweets extends the Convolutional Neural Networks state of the art approach and begins to initialize the input representations with embeddings trained on different units.

TL;DR: In this paper, a new neural network architecture was proposed to predict short-term price movements in stock markets, based on limit order book data, which can be used to predict stock market prices.

TL;DR: AMBER provides an efficient automated method for designing accurate deep learning models in genomics through the state-of-the-art neural architecture search and is illustrated to accurately discover functional genomic variants in allele-specific binding and disease heritability enrichment.

TL;DR: Minimal preprocessing of the data was required, but architecture of the network was highly constrained and specifically designed for the task, and has 1% error rate and about a 9% reject rate on zipcode digits provided by the U.S. Postal Service.

TL;DR: The authors present a time-delay neural network (TDNN) approach to phoneme recognition which is characterized by two important properties: using a three-layer arrangement of simple computing units, a hierarchy can be constructed that allows for the formation of arbitrary nonlinear decision surfaces, which the TDNN learns automatically using error backpropagation.

TL;DR: The results confirm the idea that minimizing the number of free parameters in the network enhances generalization, and show that Multtlayer constrained networks perform very well on this task when orgamzed in a hierarchical structure with shift invariant feature detectors.

TL;DR: A feed-forward network architecture for recognizing an unconstrained handwritten multi-digit string with segmentation done on the feature maps developed in the Space Displacement Neural Network rather than the input (pixel) space.