scholarly journals Targeted Gradient Descent: A Novel Method for Convolutional Neural Networks Fine-Tuning and Online-Learning

2021 ◽  
pp. 25-35
Author(s):  
Junyu Chen ◽  
Evren Asma ◽  
Chung Chan
Keyword(s):  
Neural Networks ◽  
Online Learning ◽  
Convolutional Neural Networks ◽  
Gradient Descent ◽  
Fine Tuning ◽  
Novel Method

scholarly journals Direct Quantization for Training Highly Accurate Low Bit-width Deep Neural Networks

2020 ◽  
Author(s):  
Tuan Hoang ◽  
Thanh-Toan Do ◽  
Tam V. Nguyen ◽  
Ngai-Man Cheung
Keyword(s):  
Neural Networks ◽  
Cost Function ◽  
Image Classification ◽  
Convolutional Neural Networks ◽  
Gradient Descent ◽  
Deep Neural Networks ◽  
State Of The Art ◽  
Deep Convolutional Neural Networks ◽  
Novel Method ◽  
The Cost

This paper proposes two novel techniques to train deep convolutional neural networks with low bit-width weights and activations. First, to obtain low bit-width weights, most existing methods obtain the quantized weights by performing quantization on the full-precision network weights. However, this approach would result in some mismatch: the gradient descent updates full-precision weights, but it does not update the quantized weights. To address this issue, we propose a novel method that enables direct updating of quantized weights with learnable quantization levels to minimize the cost function using gradient descent. Second, to obtain low bit-width activations, existing works consider all channels equally. However, the activation quantizers could be biased toward a few channels with high-variance. To address this issue, we propose a method to take into account the quantization errors of individual channels. With this approach, we can learn activation quantizers that minimize the quantization errors in the majority of channels. Experimental results demonstrate that our proposed method achieves state-of-the-art performance on the image classification task, using AlexNet, ResNet and MobileNetV2 architectures on CIFAR-100 and ImageNet datasets.

scholarly journals Driving Stress Detection Using Multimodal Convolutional Neural Networks with Nonlinear Representation of Short-Term Physiological Signals

Sensors ◽  
2021 ◽  
Vol 21 (7) ◽  
pp. 2381
Author(s):  
Jaewon Lee ◽  
Hyeonjeong Lee ◽  
Miyoung Shin
Keyword(s):  
Neural Networks ◽  
Convolutional Neural Networks ◽  
Traffic Accidents ◽  
Physiological Signals ◽  
Superior Performance ◽  
Short Term ◽  
Input Signals ◽  
Skin Response ◽  
Novel Method ◽  
Fully Connected

Mental stress can lead to traffic accidents by reducing a driver’s concentration or increasing fatigue while driving. In recent years, demand for methods to detect drivers’ stress in advance to prevent dangerous situations increased. Thus, we propose a novel method for detecting driving stress using nonlinear representations of short-term (30 s or less) physiological signals for multimodal convolutional neural networks (CNNs). Specifically, from hand/foot galvanic skin response (HGSR, FGSR) and heart rate (HR) short-term input signals, first, we generate corresponding two-dimensional nonlinear representations called continuous recurrence plots (Cont-RPs). Second, from the Cont-RPs, we use multimodal CNNs to automatically extract FGSR, HGSR, and HR signal representative features that can effectively differentiate between stressed and relaxed states. Lastly, we concatenate the three extracted features into one integrated representation vector, which we feed to a fully connected layer to perform classification. For the evaluation, we use a public stress dataset collected from actual driving environments. Experimental results show that the proposed method demonstrates superior performance for 30-s signals, with an overall accuracy of 95.67%, an approximately 2.5–3% improvement compared with that of previous works. Additionally, for 10-s signals, the proposed method achieves 92.33% classification accuracy, which is similar to or better than the performance of other methods using long-term signals (over 100 s).

scholarly journals Layer-Wise Compressive Training for Convolutional Neural Networks

2018 ◽  
Vol 11 (1) ◽  
pp. 7 ◽  
Author(s):  
Matteo Grimaldi ◽  
Valerio Tenace ◽  
Andrea Calimera
Keyword(s):  
Neural Networks ◽  
Convolutional Neural Networks ◽  
Gradient Descent ◽  
Computational Models ◽  
Stochastic Gradient Descent ◽  
Training Algorithm ◽  
Heuristic Rules ◽  
Human Capabilities ◽  
Model Size ◽  
Large Model

Convolutional Neural Networks (CNNs) are brain-inspired computational models designed to recognize patterns. Recent advances demonstrate that CNNs are able to achieve, and often exceed, human capabilities in many application domains. Made of several millions of parameters, even the simplest CNN shows large model size. This characteristic is a serious concern for the deployment on resource-constrained embedded-systems, where compression stages are needed to meet the stringent hardware constraints. In this paper, we introduce a novel accuracy-driven compressive training algorithm. It consists of a two-stage flow: first, layers are sorted by means of heuristic rules according to their significance; second, a modified stochastic gradient descent optimization is applied on less significant layers such that their representation is collapsed into a constrained subspace. Experimental results demonstrate that our approach achieves remarkable compression rates with low accuracy loss (<1%).

scholarly journals Performance of Fine-Tuning Convolutional Neural Networks for HEp-2 Image Classification

2020 ◽  
Vol 10 (19) ◽  
pp. 6940 ◽  
Author(s):  
Vincenzo Taormina ◽  
Donato Cascio ◽  
Leonardo Abbene ◽  
Giuseppe Raso
Keyword(s):  
Neural Networks ◽  
Convolutional Neural Networks ◽  
Fluorescence Intensity ◽  
Characteristic Curve ◽  
Performance Comparison ◽  
Fine Tuning ◽  
Tuning Method ◽  
Effective Performance ◽  
Training Modalities

The search for anti-nucleus antibodies (ANA) represents a fundamental step in the diagnosis of autoimmune diseases. The test considered the gold standard for ANA research is indirect immunofluorescence (IIF). The best substrate for ANA detection is provided by Human Epithelial type 2 (HEp-2) cells. The first phase of HEp-2 type image analysis involves the classification of fluorescence intensity in the positive/negative classes. However, the analysis of IIF images is difficult to perform and particularly dependent on the experience of the immunologist. For this reason, the interest of the scientific community in finding relevant technological solutions to the problem has been high. Deep learning, and in particular the Convolutional Neural Networks (CNNs), have demonstrated their effectiveness in the classification of biomedical images. In this work the efficacy of the CNN fine-tuning method applied to the problem of classification of fluorescence intensity in HEp-2 images was investigated. For this purpose, four of the best known pre-trained networks were analyzed (AlexNet, SqueezeNet, ResNet18, GoogLeNet). The classifying power of CNN was investigated with different training modalities; three levels of freezing weights and scratch. Performance analysis was conducted, in terms of area under the ROC (Receiver Operating Characteristic) curve (AUC) and accuracy, using a public database. The best result achieved an AUC equal to 98.6% and an accuracy of 93.9%, demonstrating an excellent ability to discriminate between the positive/negative fluorescence classes. For an effective performance comparison, the fine-tuning mode was compared to those in which CNNs are used as feature extractors, and the best configuration found was compared with other state-of-the-art works.

A novel method based on convolutional neural networks for deriving standard 12-lead ECG from serial 3-lead ECG

2019 ◽  
Vol 20 (3) ◽  
pp. 405-413 ◽  
Author(s):  
Lu-di Wang ◽  
Wei Zhou ◽  
Ying Xing ◽  
Na Liu ◽  
Mahmood Movahedipour ◽  
...  
Keyword(s):  
Neural Networks ◽  
Convolutional Neural Networks ◽  
Novel Method

scholarly journals An Efficient Algorithm for Cardiac Arrhythmia Classification Using Ensemble of Depthwise Separable Convolutional Neural Networks

2020 ◽  
Vol 10 (2) ◽  
pp. 483 ◽  
Author(s):  
Eko Ihsanto ◽  
Kalamullah Ramli ◽  
Dodi Sudiana ◽  
Teddy Surya Gunawan
Keyword(s):  
Neural Networks ◽  
Feature Extraction ◽  
Cardiac Arrhythmia ◽  
Convolutional Neural Networks ◽  
Computational Cost ◽  
Training Data ◽  
Qrs Detection ◽  
Convolutional Network ◽  
Novel Method ◽  
Electrocardiogram Ecg

Many algorithms have been developed for automated electrocardiogram (ECG) classification. Due to the non-stationary nature of the ECG signal, it is rather challenging to use traditional handcraft methods, such as time-based analysis of feature extraction and classification, to pave the way for machine learning implementation. This paper proposed a novel method, i.e., the ensemble of depthwise separable convolutional (DSC) neural networks for the classification of cardiac arrhythmia ECG beats. Using our proposed method, the four stages of ECG classification, i.e., QRS detection, preprocessing, feature extraction, and classification, were reduced to two steps only, i.e., QRS detection and classification. No preprocessing method was required while feature extraction was combined with classification. Moreover, to reduce the computational cost while maintaining its accuracy, several techniques were implemented, including All Convolutional Network (ACN), Batch Normalization (BN), and ensemble convolutional neural networks. The performance of the proposed ensemble CNNs were evaluated using the MIT-BIH arrythmia database. In the training phase, around 22% of the 110,057 beats data extracted from 48 records were utilized. Using only these 22% labeled training data, our proposed algorithm was able to classify the remaining 78% of the database into 16 classes. Furthermore, the sensitivity ( S n ), specificity ( S p ), and positive predictivity ( P p ), and accuracy ( A c c ) are 99.03%, 99.94%, 99.03%, and 99.88%, respectively. The proposed algorithm required around 180 μs, which is suitable for real time application. These results showed that our proposed method outperformed other state of the art methods.

Sign in / Sign up

Export Citation Format

Share Document