DNN model compression for IoT domain-specific hardware accelerators

E Russo, M Palesi, S Monteleone… - IEEE Internet of …, 2021 - ieeexplore.ieee.org
Machine learning techniques, particularly those based on neural networks, are always more
often used at the edge of the network by Internet of Things (IoT) nodes. Unfortunately, the …

Improving inference latency and energy of DNNs through wireless enabled multi-chip-module-based architectures and model parameters compression

G Ascia, V Catania, A Mineo… - 2020 14th IEEE/ACM …, 2020 - ieeexplore.ieee.org
Performance and energy figures of Deep Neural Network (DNN) accelerators are profoundly
affected by the communication and memory sub-system. In this paper, we make the case of …

LAMBDA: An open framework for deep neural network accelerators simulation

E Russo, M Palesi, S Monteleone… - … and other Affiliated …, 2021 - ieeexplore.ieee.org
Many tasks in the realm of recognition, mining, and synthesis are increasingly being
implemented by using machine learning approaches. In particular, deep neural networks …

The position-based compression techniques for DNN model

M Tang, E Russo, M Palesi - The Journal of Supercomputing, 2023 - Springer
In deep neural network (DNN) accelerators, it is expensive to transfer model parameters
from the main memory to the processing elements. Data movement accounts for a large …

Analyzing the Impact of DNN Hardware Accelerators-Oriented Compression Techniques on General-Purpose Low-End Boards

G Canzonieri, S Monteleone, M Palesi, E Russo… - … Conference on Mobile …, 2022 - Springer
Abstract Deep Neural Networks emerged in the last years as the most promising approach
to the smart processing of data. However, their effectiveness is still a challenge when they …

Mapping and virtual neuron assignment algorithms for MAERI accelerator

M Reshadi, SYH Mirmahaleh - The Journal of Supercomputing, 2022 - Springer
To date, some different deep learning accelerators (DLAs) have proposed to solve
challenges caused by increasing deep neural networks' layers. GPU-based systems almost …