Results 41 to 50 of about 75,210 (267)

Exploring Efficient Neural Architectures for Linguistic–Acoustic Mapping in Text-To-Speech

open access: yesApplied Sciences, 2019
Conversion from text to speech relies on the accurate mapping from linguistic to acoustic symbol sequences, for which current practice employs recurrent statistical models such as recurrent neural networks. Despite the good performance of such models (in
Santiago Pascual   +2 more
doaj   +1 more source

Segmental Recurrent Neural Networks

open access: yes, 2015
We introduce segmental recurrent neural networks (SRNNs) which define, given an input sequence, a joint probability distribution over segmentations of the input and labelings of the segments. Representations of the input segments (i.e., contiguous subsequences of the input) are computed by encoding their constituent tokens using bidirectional recurrent
Lingpeng Kong, Chris Dyer, Noah A. Smith
openaire   +2 more sources

Quaternion Recurrent Neural Networks

open access: yesCoRR, 2018
ICLR Update - Full ...
Parcollet, Titouan   +6 more
openaire   +4 more sources

Relational recurrent neural networks

open access: yesCoRR, 2018
Memory-based neural networks model temporal data by leveraging an ability to remember information for long periods. It is unclear, however, whether they also have an ability to perform complex relational reasoning with the information they remember.
Adam Santoro   +9 more
openaire   +3 more sources

Feasibility and Safety of Somato‐Cognitive Coordination Therapy for Cerebellar Ataxia Following Pediatric Brain Tumor Treatment

open access: yesPediatric Blood &Cancer, EarlyView.
ABSTRACT Background Cerebellar ataxia after pediatric brain tumor treatment can cause persistent gait, balance, and speech impairment, yet no established rehabilitation strategy exists. Somato‐cognitive coordination therapy (SCCT) is a virtual reality–guided intervention designed to promote sensorimotor integration through visually constrained reaching
Masanobu Takeuchi   +10 more
wiley   +1 more source

Dilated Recurrent Neural Networks

open access: yesCoRR, 2017
Learning with recurrent neural networks (RNNs) on long sequences is a notoriously difficult task. There are three major challenges: 1) complex dependencies, 2) vanishing and exploding gradients, and 3) efficient parallelization. In this paper, we introduce a simple yet effective RNN connection structure, the DilatedRNN, which simultaneously tackles all
Shiyu Chang   +9 more
openaire   +3 more sources

The Role of Chemotherapy in Pediatric Myoepithelial Carcinoma: A Systematic Review of the Literature

open access: yesPediatric Blood &Cancer, EarlyView.
ABSTRACT Myoepithelial carcinoma (MEC) in pediatric patients is a rare and aggressive malignancy characterized by heterogeneous morphology and variable molecular features. The optimal role of chemotherapy remains unclear. We conducted a systematic review according to PRISMA 2020 guidelines to evaluate chemotherapy in pediatric and young‐adult patients ...
Marco Salvi   +7 more
wiley   +1 more source

Linked Recurrent Neural Networks

open access: yesCoRR, 2018
Recurrent Neural Networks (RNNs) have been proven to be effective in modeling sequential data and they have been applied to boost a variety of tasks such as document classification, speech recognition and machine translation. Most of existing RNN models have been designed for sequences assumed to be identically and independently distributed (i.i.d ...
Zhiwei Wang 0001   +3 more
openaire   +2 more sources

Reciprocal control of viral infection and phosphoinositide dynamics

open access: yesFEBS Letters, EarlyView.
Phosphoinositides, although scarce, regulate key cellular processes, including membrane dynamics and signaling. Viruses exploit these lipids to support their entry, replication, assembly, and egress. The central role of phosphoinositides in infection highlights phosphoinositide metabolism as a promising antiviral target.
Marie Déborah Bancilhon, Bruno Mesmin
wiley   +1 more source

Grounded Recurrent Neural Networks

open access: yesCoRR, 2017
In this work, we present the Grounded Recurrent Neural Network (GRNN), a recurrent neural network architecture for multi-label prediction which explicitly ties labels to specific dimensions of the recurrent hidden state (we call this process "grounding"). The approach is particularly well-suited for extracting large numbers of concepts from text.
Ankit Vani   +2 more
openaire   +2 more sources

Home - About - Disclaimer - Privacy