TY - GEN
T1 - Aggressively prunable MAM²-based Deep Neural Oracle for ECG acquisition by Compressed Sensing
AU - Bich, Philippe
AU - Prono, Luciano
AU - Mangia, Mauro
AU - Pareschi, Fabio
AU - Rovatti, Riccardo
AU - Setti, Gianluca
N1 - Generated from Scopus record by KAUST IRTS on 2023-02-15
PY - 2022/1/1
Y1 - 2022/1/1
N2 - The growing interest in Internet of Things (IoT) and mobile biomedical applications is pushing the investigation on approaches that can be used to reduce the energy consumption while acquiring data. Compressed Sensing (CS) is a technique that allows to reduce the energy required for the acquisition and compression of a sparse signal, transferring the complexity to the reconstruction stage. Many works leverage the use of Deep Neural Networks (DNNs) for signal reconstruction and, assuming that also this operation has to be performed on a IoT device, it is necessary for the DNN architecture to fit in small and low-energy devices. Pruning techniques, that can reduce the size of DNNs by removing unnecessary parameters and thus decreasing storage requirements, can be of great help in this effort. In this work, a novel Multiply and Max&Min (MAM²) map-reduce paradigm trained with the vanishing contributes technique and then pruned with the activation rate method is proposed. The result is a naturally and aggressively pruned DNN layer structure. This structure is used to reduce the complexity of a DNN-based CS reconstructor and its performance is verified. As an example, MAM²-based layers still retain the baseline accuracy of the CS decoder with 94% of the parameters pruned against 25% when using classic MAC-based layers only.
AB - The growing interest in Internet of Things (IoT) and mobile biomedical applications is pushing the investigation on approaches that can be used to reduce the energy consumption while acquiring data. Compressed Sensing (CS) is a technique that allows to reduce the energy required for the acquisition and compression of a sparse signal, transferring the complexity to the reconstruction stage. Many works leverage the use of Deep Neural Networks (DNNs) for signal reconstruction and, assuming that also this operation has to be performed on a IoT device, it is necessary for the DNN architecture to fit in small and low-energy devices. Pruning techniques, that can reduce the size of DNNs by removing unnecessary parameters and thus decreasing storage requirements, can be of great help in this effort. In this work, a novel Multiply and Max&Min (MAM²) map-reduce paradigm trained with the vanishing contributes technique and then pruned with the activation rate method is proposed. The result is a naturally and aggressively pruned DNN layer structure. This structure is used to reduce the complexity of a DNN-based CS reconstructor and its performance is verified. As an example, MAM²-based layers still retain the baseline accuracy of the CS decoder with 94% of the parameters pruned against 25% when using classic MAC-based layers only.
UR - https://ieeexplore.ieee.org/document/9948676/
UR - http://www.scopus.com/inward/record.url?scp=85142929877&partnerID=8YFLogxK
U2 - 10.1109/BioCAS54905.2022.9948676
DO - 10.1109/BioCAS54905.2022.9948676
M3 - Conference contribution
SN - 9781665469173
SP - 163
EP - 167
BT - BioCAS 2022 - IEEE Biomedical Circuits and Systems Conference: Intelligent Biomedical Systems for a Better Future, Proceedings
PB - Institute of Electrical and Electronics Engineers Inc.
ER -
