Deep-neural-networks (DNNs) have significantly improved automatic speech recognition (ASR) accuracy over a range of speech scenarios. However noise-robustness is still a challenge to DNNs, where compared to clean, accuracy degrades significantly for noisy environments. Many of the current DNN-based ASR engines use log-MelSpectra features, along with features from temporal-difference in delta an...

Deep neural networks (DNNs) have set state of the art results in many machine learning and NLP tasks. However, we do not have a strong understanding of what DNN models learn. In this paper, we examine learning in DNNs through analysis of their outputs. We compare DNN performance directly to a human population, and use characteristics of individual data points such as difficulty to see how well ...

In this paper, we theoretically prove the existence of periodic solutions for a nonautonomous discrete-time neural networks by using the topological degree theory. Sufficient conditions are also obtained for the existence of an asymptotically stable periodic solution. As a special case, we obtain the existence of a fixed point to the corresponding autonomous discrete-time neural networks which ...

An important asset of using Deep Neural Networks (DNNs) for text applications is their ability to automatically engineer features. Unfortunately, DNNs usually require a lot of training data, especially for high-level semantic tasks such as community Question Answering (cQA). In this paper, we tackle the problem of data scarcity by learning the target DNN together with two auxiliary tasks in a m...

In this paper we describe the development of an accurate, smallfootprint, large vocabulary speech recognizer for mobile devices. To achieve the best recognition accuracy, state-of-the-art deep neural networks (DNNs) are adopted as acoustic models. A variety of speedup techniques for DNN score computation are used to enable real-time operation on mobile devices. To reduce the memory and disk usa...

Deep neural networks (DNNs) have achieved unprecedented performance on a wide range of complex tasks, rapidly outpacing our understanding of the nature of their solutions. This has caused a recent surge of interest in methods for rendering modern neural systems more interpretable. In this work, we propose to address the interpretability problem in modern DNNs using the rich history of problem d...

It is well known that metamodel or surrogate modeling techniques have been widely applied in engineering problems due to their higher efficiency. However, with the increase of the linearity and dimensions, it is difficult for the present popular metamodeling techniques to construct reliable metamodel and apply to more and more complicated high dimensional problems. Recently, neural networks (NN...

Recent research on deep neural networks has focused primarily on improving accuracy. For a given accuracy level, is typically possible to identify multiple DNN architectures that achieve that accuracy level. With equivalent accuracy, smaller DNN architectures offer at least three advantages: (1) Smaller DNNs require less communication across servers during distributed training. (2) Smaller DNNs...