#### Filter Results:

- Full text PDF available (8)

#### Publication Year

1987

2017

- This year (1)
- Last 5 years (4)
- Last 10 years (8)

#### Publication Type

#### Co-author

#### Journals and Conferences

#### Data Set Used

#### Key Phrases

Learn More

- Jeffrey Dean, Gregory S. Corrado, +9 authors Andrew Y. Ng
- NIPS
- 2012

Recent work in unsupervised feature learning and deep learning has shown that being able to train large models can dramatically improve performance. In this paper, we consider the problem of training a deep network with billions of parameters using tens of thousands of CPU cores. We have developed a software framework called DistBelief that can utilize… (More)

- Matthew D. Zeiler, Marc'Aurelio Ranzato, +8 authors Geoffrey E. Hinton
- 2013 IEEE International Conference on Acoustics…
- 2013

Deep neural networks have recently become the gold standard for acoustic modeling in speech recognition systems. The key computational unit of a deep network is a linear projection followed by a point-wise non-linearity, which is typically a logistic function. In this work, we show that we can improve generalization and make training of deep networks faster… (More)

Recent advances in deep learning have made the use of large, deep neural networks with tens of millions of parameters suitable for a number of applications that require real-time processing. The sheer size of these networks can represent a challenging computational burden, even for modern CPUs. For this reason, GPUs are routinely used instead to train and… (More)

- Hasim Sak, Oriol Vinyals, +4 authors Mark Z. Mao
- INTERSPEECH
- 2014

We recently showed that Long Short-Term Memory (LSTM) recurrent neural networks (RNNs) outperform state-of-the-art deep neural networks (DNNs) for large scale acoustic modeling where the models were trained with the cross-entropy (CE) criterion. It has also been shown that sequence discriminative training of DNNs initially trained with the CE criterion… (More)

- Mark Z. Mao, Robert M. Gray, Tamás Linder
- 2011 Data Compression Conference
- 2011

A vector extension of a necessary condition for asymptotically optimal stationary (sliding-block) source codes is presented. The condition implies the intuitive result that the reproduction process for an IID input must be approximately uncorrelated if the code is approximately optimal, a property previously demonstrated empirically for common examples.… (More)

The in vivo effects of a single dose of levo-praziquantel, 75 mg/kg in PEG 400, on the tegumental surface of adult S. japonicum were compared with the effects of a single dose (150 mg/kg) of the mixed isomer preparation, using scanning and transmission electron microscope. Worms were recovered from mice at 10 min, 30 min, 1 hr, 4 hr, 12 hr, 24 hr and 48 hr… (More)

- Mark Z. Mao, Vincent Vanhoucke, Brian Strope
- ICASSP
- 2005

- Larry P. Heck, Mark Z. Mao
- INTERSPEECH
- 2004

This paper presents a novel Bayesian approach to the problem of co-channel speech. The problem is formulated as the joint maximization of the a posteriori probability of the word sequence and the target speaker given the observed speech signal. It is shown that the joint probability can be expressed as the product of six terms: a likelihood score from a… (More)

- Mark Z. Mao, Robert M. Gray, Tamás Linder
- IEEE Transactions on Information Theory
- 2011

Necessary conditions for asymptotically optimal sliding-block or stationary codes for source coding and rate-constrained simulation of memoryless sources are presented and used to motivate a design technique for trellis-encoded source coding and rate-constrained simulation. The code structure has intuitive similarities to classic random coding arguments as… (More)

- Sangho Yoon, Mark Z. Mao, Mark Kalman
- ICME
- 2006

We propose a new real-time packet scheduling algorithm for streaming scalable H.264. Our algorithmmakes use of a packet importance measure, which we define, that takes into consideration transmission history, channel conditions, and the unique decoding dependencies due to the temporal wavelet encoding. Our algorithm utilizes this importance measure to… (More)