A High-Performance Accelerator for Large-Scale Convolutional Neural Networks

2017 ◽  
Author(s):  
Fan Sun ◽  
Chao Wang ◽  
Lei Gong ◽  
Chongchong Xu ◽  
Yiwei Zhang ◽  
...  
Keyword(s):  
Neural Networks ◽  
Convolutional Neural Networks ◽  
High Performance ◽  
Large Scale

scholarly journals Energy-Efficient Architecture for CNNs Inference on Heterogeneous FPGA

2019 ◽  
Vol 10 (1) ◽  
pp. 1 ◽  
Author(s):  
Fanny Spagnolo ◽  
Stefania Perri ◽  
Fabio Frustaci ◽  
Pasquale Corsonello
Keyword(s):  
Neural Networks ◽  
Embedded System ◽  
Convolutional Neural Networks ◽  
Energy Efficient ◽  
Power Efficiency ◽  
High Performance ◽  
Large Scale ◽  
Low Cost ◽  
Frame Rate ◽  
Field Programmable

Due to the huge requirements in terms of both computational and memory capabilities, implementing energy-efficient and high-performance Convolutional Neural Networks (CNNs) by exploiting embedded systems still represents a major challenge for hardware designers. This paper presents the complete design of a heterogeneous embedded system realized by using a Field-Programmable Gate Array Systems-on-Chip (SoC) and suitable to accelerate the inference of Convolutional Neural Networks in power-constrained environments, such as those related to IoT applications. The proposed architecture is validated through its exploitation in large-scale CNNs on low-cost devices. The prototype realized on a Zynq XC7Z045 device achieves a power efficiency up to 135 Gops/W. When the VGG-16 model is inferred, a frame rate up to 11.8 fps is reached.

USE OF CONVOLUTIONAL NEURAL NETWORKS FOR X-RAY IMAGE ORIENTATION DETERMINATION

2021 ◽  
Author(s):  
Sandi Baressi Šegota ◽  
◽  
Simon Lysdahlgaard ◽  
Søren Hess ◽  
Ronald Antulov
Keyword(s):  
Neural Networks ◽  
Convolutional Neural Networks ◽  
High Performance ◽  
Data Augmentation ◽  
Classification Model ◽  
Orientation Sensitivity ◽  
X Ray ◽  
Artificial Neural Network Ann ◽  
Image Orientation ◽  
Single Orientation

The fact that Artificial Intelligence (AI) based algorithms exhibit a high performance on image classification tasks has been shown many times. Still, certain issues exist with the application of machine learning (ML) artificial neural network (ANN) algorithms. The best known is the need for a large amount of statistically varied data, which can be addressed with expanded collection or data augmentation. Other issues are also present. Convolutional neural networks (CNNs) show extremely high performance on image-shaped data. Despite their performance, CNNs exhibit a large issue which is the sensitivity to image orientation. Previous research shows that varying the orientation of images may greatly lower the performance of the trained CNN. This is especially problematic in certain applications, such as X-ray radiography, an example of which is presented here. Previous research shows that the performance of CNNs is higher when used on images in a single orientation (left or right), as opposed to the combination of both. This means that the data needs to be differentiated before it enters the classification model. In this paper, the CNN-based model for differentiation between left and right-oriented images is presented. Multiple CNNs are trained and tested, with the highest performing being the VGG16 architecture which achieved an Accuracy of 0.99 (+/- 0.01), and an AUC of 0.98 (+/- 0.01). These results show that CNNs can be used to address the issue of orientation sensitivity by splitting the data in advance of being used in classification models.

scholarly journals MS2CNN: predicting MS/MS spectrum based on protein sequence using deep convolutional neural networks

BMC Genomics ◽  
2019 ◽  
Vol 20 (S9) ◽  
Author(s):  
Yang-Ming Lin ◽  
Ching-Tai Chen ◽  
Jia-Ming Chang
Keyword(s):  
Neural Networks ◽  
Deep Learning ◽  
Convolutional Neural Networks ◽  
Large Scale ◽  
Learning Algorithm ◽  
Pearson Correlation ◽  
Tandem Mass ◽  
Spectral Library ◽  
Deep Convolutional Neural Networks ◽  
Library Search

Abstract Background Tandem mass spectrometry allows biologists to identify and quantify protein samples in the form of digested peptide sequences. When performing peptide identification, spectral library search is more sensitive than traditional database search but is limited to peptides that have been previously identified. An accurate tandem mass spectrum prediction tool is thus crucial in expanding the peptide space and increasing the coverage of spectral library search. Results We propose MS2CNN, a non-linear regression model based on deep convolutional neural networks, a deep learning algorithm. The features for our model are amino acid composition, predicted secondary structure, and physical-chemical features such as isoelectric point, aromaticity, helicity, hydrophobicity, and basicity. MS2CNN was trained with five-fold cross validation on a three-way data split on the large-scale human HCD MS2 dataset of Orbitrap LC-MS/MS downloaded from the National Institute of Standards and Technology. It was then evaluated on a publicly available independent test dataset of human HeLa cell lysate from LC-MS experiments. On average, our model shows better cosine similarity and Pearson correlation coefficient (0.690 and 0.632) than MS2PIP (0.647 and 0.601) and is comparable with pDeep (0.692 and 0.642). Notably, for the more complex MS2 spectra of 3+ peptides, MS2PIP is significantly better than both MS2PIP and pDeep. Conclusions We showed that MS2CNN outperforms MS2PIP for 2+ and 3+ peptides and pDeep for 3+ peptides. This implies that MS2CNN, the proposed convolutional neural network model, generates highly accurate MS2 spectra for LC-MS/MS experiments using Orbitrap machines, which can be of great help in protein and peptide identifications. The results suggest that incorporating more data for deep learning model may improve performance.

Sign in / Sign up

Export Citation Format

Share Document