Notices by arXiv Statistics (arxivstats@qoto.org)

arXiv Statistics (arxivstats@qoto.org)'s status on Wednesday, 25-Dec-2019 03:20:07 UTC arXiv Statistics

power-law nonlinearity with maximally uniform distribution criterion for improved neural network training in automatic speech recognition. (arXiv:1912.11041v1 [eess.AS]) http://arxiv.org/abs/1912.11041
In conversation Wednesday, 25-Dec-2019 03:20:07 UTC from qoto.org permalink
Attachments
1. power-law nonlinearity with maximally uniform distribution criterion for improved neural network training in automatic speech recognition
  
  In this paper, we describe the Maximum Uniformity of Distribution (MUD) algorithm with the power-law nonlinearity. In this approach, we hypothesize that neural network training will become more stable if feature distribution is not too much skewed. We propose two different types of MUD approaches: power function-based MUD and histogram-based MUD. In these approaches, we first obtain the mel filterbank coefficients and apply nonlinearity functions for each filterbank channel. With the power function-based MUD, we apply a power-function based nonlinearity where power function coefficients are chosen to maximize the likelihood assuming that nonlinearity outputs follow the uniform distribution. With the histogram-based MUD, the empirical Cumulative Density Function (CDF) from the training database is employed to transform the original distribution into a uniform distribution. In MUD processing, we do not use any prior knowledge (e.g. logarithmic relation) about the energy of the incoming signal and the perceived intensity by a human. Experimental results using an end-to-end speech recognition system demonstrate that power-function based MUD shows better result than the conventional Mel Filterbank Cepstral Coefficients (MFCCs). On the LibriSpeech database, we could achieve 4.02 % WER on test-clean and 13.34 % WER on test-other without using any Language Models (LMs). The major contribution of this work is that we developed a new algorithm for designing the compressive nonlinearity in a data-driven way, which is much more flexible than the previous approaches and may be extended to other domains as well.
arXiv Statistics (arxivstats@qoto.org)'s status on Wednesday, 13-Nov-2019 03:20:06 UTC arXiv Statistics

Geometry-Aware Neural Rendering. (arXiv:1911.04554v1 [cs.CV]) http://arxiv.org/abs/1911.04554
In conversation Wednesday, 13-Nov-2019 03:20:06 UTC from qoto.org permalink
Attachments
1. Geometry-Aware Neural Rendering
  
  Understanding the 3-dimensional structure of the world is a core challenge in computer vision and robotics. Neural rendering approaches learn an implicit 3D model by predicting what a camera would see from an arbitrary viewpoint. We extend existing neural rendering to more complex, higher dimensional scenes than previously possible. We propose Epipolar Cross Attention (ECA), an attention mechanism that leverages the geometry of the scene to perform efficient non-local operations, requiring only $O(n)$ comparisons per spatial dimension instead of $O(n^2)$. We introduce three new simulated datasets inspired by real-world robotics and demonstrate that ECA significantly improves the quantitative and qualitative performance of Generative Query Networks (GQN).
arXiv Statistics (arxivstats@qoto.org)'s status on Monday, 02-Sep-2019 03:15:27 UTC arXiv Statistics

Deep Learning for Estimating Synaptic Health of Primary Neuronal Cell Culture. (arXiv:1908.11399v1 [eess.IV]) http://arxiv.org/abs/1908.11399
In conversation Monday, 02-Sep-2019 03:15:27 UTC from qoto.org permalink
Attachments
1. Deep Learning for Estimating Synaptic Health of Primary Neuronal Cell Culture
arXiv Statistics (arxivstats@qoto.org)'s status on Monday, 02-Sep-2019 03:15:26 UTC arXiv Statistics

Translating Mathematical Formula Images to LaTeX Sequences Using Deep Neural Networks with Sequence-level Training. (arXiv:1908.11415v1 [cs.LG]) http://arxiv.org/abs/1908.11415
In conversation Monday, 02-Sep-2019 03:15:26 UTC from qoto.org permalink
Attachments
1. Translating Mathematical Formula Images to LaTeX Sequences Using Deep Neural Networks with Sequence-level Training

Public

Notices by arXiv Statistics (arxivstats@qoto.org)

User actions

Following 0

Followers 0

Groups 0

Statistics

Feeds