CPU-accelerator co-scheduling for CNN acceleration at the edge

Y Kim, J Kong, A Munir - IEEE Access, 2020 - ieeexplore.ieee.org
Convolutional neural networks (CNNs) are widely deployed for many artificial intelligence
(AI) applications, such as object detection and image classification. Due to the burgeoning …

A systematic evaluation of recurrent neural network models for edge intelligence and human activity recognition applications

VS Lalapura, VR Bhimavarapu, J Amudha… - Algorithms, 2024 - mdpi.com
The Recurrent Neural Networks (RNNs) are an essential class of supervised learning
algorithms. Complex tasks like speech recognition, machine translation, sentiment …

Flydeling: Streamlined performance models for hardware acceleration of CNNs through system identification

W Carballo-Hernández, M Pelcat… - ACM Transactions on …, 2023 - dl.acm.org
The introduction of deep learning algorithms, such as Convolutional Neural Networks
(CNNs) in many near-sensor embedded systems, opens new challenges in terms of energy …

The effects of partitioning strategies on energy consumption in distributed cnn inference at the edge

E Tang, X Guo, T Stefanov - arXiv preprint arXiv:2210.08392, 2022 - arxiv.org
Nowadays, many AI applications utilizing resource-constrained edge devices (eg, small
mobile robots, tiny IoT devices, etc.) require Convolutional Neural Network (CNN) inference …

Automatic CNN Model Partitioning for GPU/FPGA-based Embedded Heterogeneous Accelerators using Geometric Programming

W Carballo-Hernández, M Pelcat, F Berry - Journal of Signal Processing …, 2023 - Springer
Abstract Graphics Processing Unit (GPU), dedicated Application Specific Integrated Circuit
(ASIC) and Field Programmable Gate Array (FPGA) accelerators are currently platforms of …

Deep Learning Process Integration on Heterogeneous GPU/FPGA Embedded Platforms

WC Hernandez - 2022 - theses.hal.science
Deep Learning (DL) algorithm deployment on edge devices, such as Convolutional Neural
Network (CNN) inference, has established a high computing demand on devices with limited …