14.3 A 28nm SoC with a 1.2GHz 568nJ/prediction sparse deep-neural-network engine with >0.1 timing error rate tolerance for IoT applications

2017 ◽  
Author(s):  
Paul N. Whatmough ◽  
Sae Kyu Lee ◽  
Hyunkwang Lee ◽  
Saketh Rama ◽  
David Brooks ◽  
...  
Keyword(s):  
Neural Network ◽  
Error Rate ◽  
Deep Neural Network ◽  
Timing Error ◽  
Iot Applications

DNN Engine: A 28-nm Timing-Error Tolerant Sparse Deep Neural Network Processor for IoT Applications

2018 ◽  
Vol 53 (9) ◽  
pp. 2722-2731 ◽  
Author(s):  
Paul N. Whatmough ◽  
Sae Kyu Lee ◽  
David Brooks ◽  
Gu-Yeon Wei
Keyword(s):  
Neural Network ◽  
Deep Neural Network ◽  
Network Processor ◽  
Timing Error ◽  
Iot Applications ◽  
28 Nm

scholarly journals Tri-net for Semi-Supervised Deep Learning

2018 ◽  
Author(s):  
Dong-Dong Chen ◽  
Wei Wang ◽  
Wei Gao ◽  
Zhi-Hua Zhou
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Deep Learning ◽  
Error Rate ◽  
Deep Neural Network ◽  
Deep Neural Networks ◽  
State Of The Art ◽  
Fine Tuning ◽  
Learning Methods ◽  
Model Initialization

Deep neural networks have witnessed great successes in various real applications, but it requires a large number of labeled data for training. In this paper, we propose tri-net, a deep neural network which is able to use massive unlabeled data to help learning with limited labeled data. We consider model initialization, diversity augmentation and pseudo-label editing simultaneously. In our work, we utilize output smearing to initialize modules, use fine-tuning on labeled data to augment diversity and eliminate unstable pseudo-labels to alleviate the influence of suspicious pseudo-labeled data. Experiments show that our method achieves the best performance in comparison with state-of-the-art semi-supervised deep learning methods. In particular, it achieves 8.30% error rate on CIFAR-10 by using only 4000 labeled examples.

scholarly journals Driver’s Seat Belt Detection Using CNN

2021 ◽  
Vol 12 (5) ◽  
pp. 776-785
Author(s):  
DS Bhupal Naik, G Sai Lakshmi, V Ramakrishna Sajja, D Venkatesulu,J Nageswara Rao
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Error Rate ◽  
High Speed ◽  
Deep Neural Network ◽  
Seat Belt ◽  
Transportation System ◽  
Convolution Neural Network ◽  
Classification Algorithms ◽  
Convolution Neural Networks

Seat belt detection is one of the necessary task which are required in transportation system to reduce accidents due to abrupt stop or high speed accident with other vehicles. In this paper, a technique is proposed to detect whether the driver wears seat belt or not by using convolution neural networks. Convolution Neural Network is nothing but deep Neural Network. ConvNet automatically collects features using filters or kernels from images without human involvement to classify the output images. Compared to different classification algorithms, preprocessing required in ConvNet is least. In this proposed method, first ConvNet is built and trained using Seatbelt dataset of both standard and non-standard. ConvNet learns the features from the images of seat belt dataset and performed better with an accuracy of 91.4% over SVM with 87.17% and an error rate of 8.55% when compared with SVM with 12.83% in case of standard dataset.

scholarly journals Deep-neural network approaches for speech recognition with heterogeneous groups of speakers including children

2016 ◽  
Vol 23 (3) ◽  
pp. 325-350 ◽  
Author(s):  
ROMAIN SERIZEL ◽  
DIEGO GIULIANI
Keyword(s):  
Neural Network ◽  
Speech Recognition ◽  
Error Rate ◽  
Deep Neural Network ◽  
Vocal Tract ◽  
Rate Performance ◽  
Posterior Probabilities ◽  
Mel Frequency Cepstral Coefficients ◽  
Heterogeneous Groups ◽  
Cepstral Coefficients

AbstractThis paper introduces deep neural network (DNN)–hidden Markov model (HMM)-based methods to tackle speech recognition in heterogeneous groups of speakers including children. We target three speaker groups consisting of children, adult males and adult females. Two different kind of approaches are introduced here: approaches based on DNN adaptation and approaches relying on vocal-tract length normalisation (VTLN). First, the recent approach that consists in adapting a general DNN to domain/language specific data is extended to target age/gender groups in the context of DNN–HMM. Then, VTLN is investigated by training a DNN–HMM system by using either mel frequency cepstral coefficients normalised with standard VTLN or mel frequency cepstral coefficients derived acoustic features combined with the posterior probabilities of the VTLN warping factors. In this later, novel, approach the posterior probabilities of the warping factors are obtained with a separate DNN and the decoding can be operated in a single pass when the VTLN approach requires two decoding passes. Finally, the different approaches presented here are combined to take advantage of their complementarity. The combination of several approaches is shown to improve the baseline phone error rate performance by thirty per cent to thirty-five per cent relative and the baseline word error rate performance by about ten per cent relative.

scholarly journals High-Quality Genomes of Nanopore Sequencing by Homologous Polishing

2020 ◽  
Author(s):  
Yao-Ting Huang ◽  
Po-Yu Liu ◽  
Pei-Wen Shih
Keyword(s):  
Neural Network ◽  
Error Correction ◽  
Error Rate ◽  
Deep Neural Network ◽  
State Of The Art ◽  
Systematic Errors ◽  
Nanopore Sequencing ◽  
High Quality ◽  
Microbial Genomes ◽  
Homologous Sequences

AbstractNanopore sequencing has been widely used for reconstruction of a variety of microbial genomes. Owing to the higher error rate, the assembled genome requires further error correction. Existing methods erase many of these errors via deep neural network trained from Nanopore reads. However, quite a few systematic errors are still left on the genome. This paper proposed a new model trained from homologous sequences extracted from closely-related genomes, which provides valuable features missed in Nanopore reads. The developed program (called Homopolish) outperforms the state-of-the-art Racon/Medaka and MarginPolish/HELEN pipelines in metagenomic and isolates of bacteria, viruses and fungi. When Homopolish is combined with Medaka or with HELEN, the genomes quality can exceed Q50 on R9.4 flowcells. The genome quality can be also improved on R10.3 flowcells (Q50-Q90). We proved that Nanopore-only sequencing can now produce high-quality genomes without the need of Illumina hybrid sequencing.

Optimizing Error Rate in Intrusion Detection System Using Artificial Neural Network Algorithm

2018 ◽  
Vol 6 (9) ◽  
pp. 152
Author(s):  
S. Vijaya Rani ◽  
G. N. K. Suresh Babu
Keyword(s):  
Neural Network ◽  
Artificial Neural Network ◽  
Intrusion Detection ◽  
Error Rate ◽  
Learning Process ◽  
Nearest Neighbor ◽  
Detection System ◽  
Support Vector ◽  
K Nearest Neighbor ◽  
Artificial Neural

The illegal hackers  penetrate the servers and networks of corporate and financial institutions to gain money and extract vital information. The hacking varies from one computing system to many system. They gain access by sending malicious packets in the network through virus, worms, Trojan horses etc. The hackers scan a network through various tools and collect information of network and host. Hence it is very much essential to detect the attacks as they enter into a network. The methods  available for intrusion detection are Naive Bayes, Decision tree, Support Vector Machine, K-Nearest Neighbor, Artificial Neural Networks. A neural network consists of processing units in complex manner and able to store information and make it functional for use. It acts like human brain and takes knowledge from the environment through training and learning process. Many algorithms are available for learning process This work carry out research on analysis of malicious packets and predicting the error rate in detection of injured packets through artificial neural network algorithms.

Sign in / Sign up

Export Citation Format

Share Document