Blackthorn: Latency Estimation Framework for CNNs on Embedded Nvidia Platforms

Lechner, Martin; Jantsch, Axel

doi:10.1109/access.2021.3101936

Cited by 9 publications

(1 citation statement)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…These achievements have encouraged researchers to apply deep learning methods for various complex tasks like object detection, object classification, human detection, etc. [16,20,22].…”

Section: Introductionmentioning

confidence: 99%

Hardware Response and Performance Analysis of Multicore Computing Systems for Deep Learning Algorithms

Kumar

Singh

2022

Cybernetics and Information Technologies

View full text Add to dashboard Cite

With the advancement in technological world, the technologies like Artificial Intelligence (AI), Machine Learning (ML), and Deep Learning (DL) are gaining more popularity in many applications of computer vision like object classification, object detection, Human detection, etc., ML and DL approaches are highly compute-intensive and require advanced computational resources for implementation. Multicore CPUs and GPUs with a large number of dedicated processor cores are typically the more prevailing and effective solutions for the high computational need. In this manuscript, we have come up with an analysis of how these multicore hardware technologies respond to DL algorithms. A Convolutional Neural Network (CNN) model have been trained for three different classification problems using three different datasets. All these experimentations have been performed on three different computational resources, i.e., Raspberry Pi, Nvidia Jetson Nano Board, & desktop computer. Results are derived for performance analysis in terms of classification accuracy and hardware response for each hardware configuration.

show abstract

“…These achievements have encouraged researchers to apply deep learning methods for various complex tasks like object detection, object classification, human detection, etc. [16,20,22].…”

Section: Introductionmentioning

confidence: 99%

Hardware Response and Performance Analysis of Multicore Computing Systems for Deep Learning Algorithms

Kumar

Singh

2022

Cybernetics and Information Technologies

View full text Add to dashboard Cite

show abstract

Hardware-Aware Evolutionary Explainable Filter Pruning for Convolutional Neural Networks

Heidorn,

Sabih,

Meyerhöfer

et al. 2024

Int J Parallel Prog

View full text Add to dashboard Cite

Filter pruning of convolutional neural networks (CNNs) is a common technique to effectively reduce the memory footprint, the number of arithmetic operations, and, consequently, inference time. Recent pruning approaches also consider the targeted device (i.e., graphics processing units) for CNN deployment to reduce the actual inference time. However, simple metrics, such as the $$\ell ^1$$ ℓ 1 -norm, are used for deciding which filters to prune. In this work, we propose a hardware-aware technique to explore the vast multi-objective design space of possible filter pruning configurations. Our approach incorporates not only the targeted device but also techniques from explainable artificial intelligence for ranking and deciding which filters to prune. For each layer, the number of filters to be pruned is optimized with the objective of minimizing the inference time and the error rate of the CNN. Experimental results show that our approach can speed up inference time by 1.40× and 1.30× for VGG-16 on the CIFAR-10 dataset and ResNet-18 on the ILSVRC-2012 dataset, respectively, compared to the state-of-the-art ABCPruner.

show abstract

SLAPP: Subgraph-level attention-based performance prediction for deep learning models

Wang,

Yang,

et al. 2024

Neural Networks

View full text Add to dashboard Cite

Blackthorn: Latency Estimation Framework for CNNs on Embedded Nvidia Platforms

Cited by 9 publications

References 21 publications

Hardware Response and Performance Analysis of Multicore Computing Systems for Deep Learning Algorithms

Hardware Response and Performance Analysis of Multicore Computing Systems for Deep Learning Algorithms

Hardware-Aware Evolutionary Explainable Filter Pruning for Convolutional Neural Networks

SLAPP: Subgraph-level attention-based performance prediction for deep learning models

Contact Info

Product

Resources

About