TY - JOUR
T1 - BioNetExplorer
T2 - Architecture-Space Exploration of Biosignal Processing Deep Neural Networks for Wearables
AU - Prabakaran, Bharath Srinivas
AU - Akhtar, Asima
AU - Rehman, Semeen
AU - Hasan, Osman
AU - Shafique, Muhammad
N1 - Funding Information:
This work was supported by the Doctoral College Resilient Embedded Systems, which is run jointly by the TU Wien s Faculty of Informatics and the UAS Technikum Wien.
Funding Information:
Manuscript received September 7, 2020; revised December 27, 2020 and January 24, 2021; accepted February 28, 2021. Date of publication March 12, 2021; date of current version August 24, 2021. This work was supported by the Doctoral College Resilient Embedded Systems, which is run jointly by the TU Wien’s Faculty of Informatics and the UAS Technikum Wien. (Bharath Srinivas Prabakaran and Asima Akhtar contributed equally to this work.) (Corresponding author: Bharath Srinivas Prabakaran.) Bharath Srinivas Prabakaran is with the Institute of Computer Engineering, Technische Universität Wien, 1040 Vienna, Austria (e-mail: bharath.prabakaran@tuwien.ac.at;).
Publisher Copyright:
© 2014 IEEE.
PY - 2021/9/1
Y1 - 2021/9/1
N2 - Deep learning (DL) has been shown to be highly effective in solving various problems across numerous applications and domains, such as autonomous driving and image recognition. Due to the advent of DL, plenty of research works have explored the applicability of DL, more specifically deep neural networks (DNNs), to solve pattern recognition and computer vision challenges. More recently, researchers have focused on the topic of automated generation and exploration of DNN architectures, which tend to mostly focus on image recognition or visual data sets, primarily, due to the computer vision-related DL advancements. In this work, we propose the BioNetExplorer framework to systematically generate and explore multiple DNN architectures for biosignal processing in wearable devices. Our framework varies key neural architecture parameters to search for an embedded DNN architecture with a low hardware overhead, which can be deployed in wearable edge devices to analyze the biosignal data and to extract the relevant information, such as arrhythmia and seizure. Furthermore, BioNetExplorer reduces the exploration time by deploying genetic algorithms, such as NSGA-II, SPEA-2, etc. Our framework also enables the hardware-aware DNN architecture search by imposing user requirements and hardware constraints (storage, FLOPs, etc.) during the exploration stage, thereby limiting the number of networks explored. Moreover, BioNetExplorer can also be used to search for DNNs based on the user-required output classes; for instance, a user might require a specific output class, attributed toward ventricular fibrillation, due to genetic predisposition or a preexisting heart condition. The use of genetic algorithms reduces the exploration time, on average, by 9\times , compared to exhaustive exploration. We are successful in identifying Pareto-optimal designs, which can reduce the storage overhead of DNN by \sim 30 MB for a quality loss of less than 0.5%. To enable low-cost embedded DNNs, BioNetExplorer also employs different model compression techniques to further reduce the storage overhead of the network by up to 53\times for a quality loss of < 0.2\%.
AB - Deep learning (DL) has been shown to be highly effective in solving various problems across numerous applications and domains, such as autonomous driving and image recognition. Due to the advent of DL, plenty of research works have explored the applicability of DL, more specifically deep neural networks (DNNs), to solve pattern recognition and computer vision challenges. More recently, researchers have focused on the topic of automated generation and exploration of DNN architectures, which tend to mostly focus on image recognition or visual data sets, primarily, due to the computer vision-related DL advancements. In this work, we propose the BioNetExplorer framework to systematically generate and explore multiple DNN architectures for biosignal processing in wearable devices. Our framework varies key neural architecture parameters to search for an embedded DNN architecture with a low hardware overhead, which can be deployed in wearable edge devices to analyze the biosignal data and to extract the relevant information, such as arrhythmia and seizure. Furthermore, BioNetExplorer reduces the exploration time by deploying genetic algorithms, such as NSGA-II, SPEA-2, etc. Our framework also enables the hardware-aware DNN architecture search by imposing user requirements and hardware constraints (storage, FLOPs, etc.) during the exploration stage, thereby limiting the number of networks explored. Moreover, BioNetExplorer can also be used to search for DNNs based on the user-required output classes; for instance, a user might require a specific output class, attributed toward ventricular fibrillation, due to genetic predisposition or a preexisting heart condition. The use of genetic algorithms reduces the exploration time, on average, by 9\times , compared to exhaustive exploration. We are successful in identifying Pareto-optimal designs, which can reduce the storage overhead of DNN by \sim 30 MB for a quality loss of less than 0.5%. To enable low-cost embedded DNNs, BioNetExplorer also employs different model compression techniques to further reduce the storage overhead of the network by up to 53\times for a quality loss of < 0.2\%.
KW - Bio-signals
KW - convolution
KW - deep neural networks (DNNs)
KW - efficiency
KW - embedded systems
KW - exploration
KW - healthcare
KW - long short-term memory (LSTM)
KW - performance
KW - wearables
UR - http://www.scopus.com/inward/record.url?scp=85102681138&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85102681138&partnerID=8YFLogxK
U2 - 10.1109/JIOT.2021.3065815
DO - 10.1109/JIOT.2021.3065815
M3 - Article
AN - SCOPUS:85102681138
VL - 8
SP - 13251
EP - 13265
JO - IEEE Internet of Things Journal
JF - IEEE Internet of Things Journal
SN - 2327-4662
IS - 17
M1 - 9377449
ER -