Time-resolved correspondences between deep neural network layers and EEG measurements in object processing

Background: Deep neural networks have become the state of the art technology for real- world classification tasks due to their ability to learn better feature representations at each layer. However, the added accuracy that is associated with the deeper layers comes at a huge cost of computation, energy and added latency. Objective: The implementations of such architectures in resource constraint IoT devices are computationally prohibitive due to its computational and memory requirements. These factors are particularly severe in IoT domain. In this paper, we propose the Adaptive Deep Neural Network (ADNN) which gets split across the compute hierarchical layers i.e. edge, fog and cloud with all splits having one or more exit locations. Methods: At every location, the data sample adaptively chooses to exit from the NN (based on confidence criteria) or get fed into deeper layers housed across different compute layers. Design of ADNN, an adaptive deep neural network which results in fast and energy- efficient decision making (inference). : Joint optimization of all the exit points in ADNN such that the overall loss is minimized. Results: Experiments on MNIST dataset show that 41.9% of samples exit at the edge location (correctly classified) and 49.7% of samples exit at fog layer. Similar results are obtained on fashion MNIST dataset with only 19.4% of the samples requiring the entire neural network layers. With this architecture, most of the data samples are locally processed and classified while maintaining the classification accuracy and also keeping in check the communication, energy and latency requirements for time sensitive IoT applications. Conclusion: We investigated the approach of distributing the layers of the deep neural network across edge, fog and the cloud computing devices wherein data samples adaptively choose the exit points to classify themselves based on the confidence criteria (threshold). The results show that the majority of the data samples are classified within the private network of the user (edge, fog) while only a few samples require the entire layers of ADNN for classification.

Download Full-text

Developing Deep Neural Network-based Denoising Techniques for Time-Resolved In Situ TEM of Catalyst Nanoparticles

Microscopy and Microanalysis ◽

10.1017/s1431927621001513 ◽

2021 ◽

Vol 27 (S1) ◽

pp. 262-264

Author(s):

Joshua Vincent ◽

Sreyas Mohan ◽

Ramon Manzorro ◽

Binh Tang ◽

Dev Sheth ◽

...

Keyword(s):

Neural Network ◽

Deep Neural Network ◽

In Situ Tem ◽

Catalyst Nanoparticles ◽

Time Resolved

Download Full-text

Time-Resolved Correspondences Between a Deep Feed-Forward Neural Network and Human Object Processing: EEG Measurements

10.32470/ccn.2019.1082-0 ◽

2019 ◽

Author(s):

Nathan Kong ◽

Blair Kaneshiro ◽

Anthony Norcia

Keyword(s):

Neural Network ◽

Feed Forward Neural Network ◽

Object Processing ◽

Time Resolved ◽

Feed Forward ◽

Human Object

Download Full-text

Towards Supporting IoT System Designers in Edge Computing Deployment Decisions

10.24834/isbn.9789178771592 ◽

2021 ◽

Author(s):

◽

Majid Ashouri

Keyword(s):

Neural Network ◽

Deep Neural Network ◽

Edge Computing ◽

Use Case ◽

Bandwidth Utilization ◽

Quality Requirements ◽

Network Layers ◽

Supporting Tool ◽

Computational Resources ◽

And Storage

The rapidly evolving Internet of Things (IoT) systems demands addressing new requirements. This particularly needs efficient deployment of IoT systems to meet the quality requirements such as latency, energy consumption, privacy, and bandwidth utilization. The increasing availability of computational resources close to the edge has prompted the idea of using these for distributed computing and storage, known as edge computing. Edge computing may help and complement cloud computing to facilitate deployment of IoT systems and improve their quality. However, deciding where to deploy the various application components is not a straightforward task, and IoT system designer should be supported for the decision. To support the designers, in this thesis we focused on the system qualities, and aimed for three main contributions. First, by reviewing the literature, we identified the relevant and most used qualities and metrics. Moreover, to analyse how computer simulation can be used as a supporting tool, we investigated the edge computing simulators, and in particular the metrics they provide for modeling and analyzing IoT systems in edge computing. Finally, we introduced a method to represent how multiple qualities can be considered in the decision. In particular, we considered distributing Deep Neural Network layers as a use case and raked the deployment options by measuring the relevant metrics via simulation.

Download Full-text

Unravelling the black box of machine learning for atrial fibrillation forecast: role of heart rate variability and of premature beats

European Heart Journal ◽

10.1093/ehjci/ehaa946.0671 ◽

2020 ◽

Vol 41 (Supplement_2) ◽

Author(s):

J.M Gregoire ◽

C Gilon ◽

S Carlier ◽

H Bersini

Keyword(s):

Neural Network ◽

Machine Learning ◽

Atrial Fibrillation ◽

Heart Rate ◽

Heart Rate Variability ◽

Deep Neural Network ◽

Black Box ◽

Rr Intervals ◽

Network Layers ◽

Premature Beats

Abstract Background The identification of patients still in sinus rhythm who will present one month later an atrial fibrillation episode is possible using machine learning (ML) techniques. However, these new ML algorithms do not provide any relevant information about the underlying pathophysiology. Purpose To compare the predictive performance for forecasting AF between a machine learning algorithm and other parameters whose pathophysiological mechanisms are known to play a role in the triggering of arrhythmias (i.e. the count of premature beats (PB) and heart rate variability (HRV) parameters) Material and methods We conducted a retrospective study from an outpatient clinic. 10484 Holter ECG recordings were screened. 250 analysable AF onsets were labelled. We developed a deep neural network model composed of convolutional neural network layers and bidirectional gated recurrent units as recurrent neural network layers that was trained for the forecast of paroxysmal AF episodes, using RR intervals variations. This model works like a black box. For comparison purposes, we used a “random forest” (RF) model of ML to obtain forecast results using HRV parameters with and without PB. This model allows the evaluation of the relevance of HRV parameters and of PB used for the forecast. We calculated the area under the curve of the receiving operating characteristic curve for the different time windows counted in RR intervals before the AF onset. Results As shown in the table, the forecasting value of the deep neural network model (ML) was not superior to the random forest algorithm. Prediction value of both decreased when analyzing the RR intervals further away from the onset of AF Conclusions These results suggest that HRV plays a predominant role in triggering AF episodes and that premature beats could add minor information. Moreover, the closer the window from AF onset, the better the accuracy, regardless of the method used. Such detection algorithms once implemented in pacemakers, might prove useful to prevent AF onset by changing pacing sequence while patients would still be in sinus rhythm, however this remains to be demonstrated Funding Acknowledgement Type of funding source: None

Download Full-text