FracBNN: Accurate and FPGA-efficient binary neural networks with fractional activations

Y Zhang, J Pan, X Liu, H Chen, D Chen… - The 2021 ACM/SIGDA …, 2021 - dl.acm.org
Binary neural networks (BNNs) have 1-bit weights and activations. Such networks are well
suited for FPGAs, as their dominant computations are bitwise arithmetic and the memory …

A survey on the optimization of neural network accelerators for micro-ai on-device inference

AN Mazumder, J Meng, HA Rashid… - IEEE Journal on …, 2021 - ieeexplore.ieee.org
Deep neural networks (DNNs) are being prototyped for a variety of artificial intelligence (AI)
tasks including computer vision, data analytics, robotics, etc. The efficacy of DNNs coincides …

Enabling design methodologies and future trends for edge AI: Specialization and codesign

C Hao, J Dotzel, J Xiong, L Benini, Z Zhang… - IEEE Design & …, 2021 - ieeexplore.ieee.org
This work is an introduction and a survey for the Special Issue on Machine Intelligence at the
Edge. The authors argue that workloads that were formerly performed in the cloud are …

VecQ: Minimal loss DNN model compression with vectorized weight quantization

C Gong, Y Chen, Y Lu, T Li, C Hao… - IEEE Transactions on …, 2020 - ieeexplore.ieee.org
Quantization has been proven to be an effective method for reducing the computing and/or
storage cost of DNNs. However, the trade-off between the quantization bitwidth and final …

Fpga-based deep learning inference accelerators: Where are we standing?

A Nechi, L Groth, S Mulhem, F Merchant… - ACM Transactions on …, 2023 - dl.acm.org
Recently, artificial intelligence applications have become part of almost all emerging
technologies around us. Neural networks, in particular, have shown significant advantages …

Tas: ternarized neural architecture search for resource-constrained edge devices

M Loni, H Mousavi, M Riazati… - … , Automation & Test …, 2022 - ieeexplore.ieee.org
Ternary Neural Networks (TNNs) compress network weights and activation functions into 2-
bit representation resulting in remarkable network compression and energy efficiency …

[HTML][HTML] Benchmarking edge computing devices for grape bunches and trunks detection using accelerated object detection single shot multibox deep learning models

SC Magalhães, FN dos Santos, P Machado… - … Applications of Artificial …, 2023 - Elsevier
Purpose: Visual perception enables robots to perceive the environment. Visual data is
processed using computer vision algorithms that are usually time-expensive and require …

WinoCNN: Kernel sharing Winograd systolic array for efficient convolutional neural network acceleration on FPGAs

X Liu, Y Chen, C Hao, A Dhar… - 2021 IEEE 32nd …, 2021 - ieeexplore.ieee.org
The combination of Winograd's algorithm and systolic array architecture has demonstrated
the capability of improving DSP efficiency in accelerating convolutional neural networks …

Algorithm/Accelerator co-design and co-search for edge AI

X Zhang, Y Li, J Pan, D Chen - IEEE Transactions on Circuits …, 2022 - ieeexplore.ieee.org
The world has seen the great success of deep neural networks (DNNs) in a massive number
of artificial intelligence (AI) applications. However, developing high-quality AI services to …

Qs-nas: Optimally quantized scaled architecture search to enable efficient on-device micro-ai

M Hosseini, T Mohsenin - … on Emerging and Selected Topics in …, 2021 - ieeexplore.ieee.org
Because of their simple hardware requirements, low bitwidth neural networks (NN) have
gained significant attention over the recent years, and have been extensively employed in …