Topics In Demand
Notification
New

No notification found.

TinyML Optimization Approaches for Deployment on Edge Devices

September 24, 2024 448 0 Machine Learning

TinyML Optimization Approaches for Deployment on Edge Devices

The demand for applications deploying deep learning models in resource-constrained environments is on the rise today, driven by the need for low latency and real-time inference. However, many deep learning models are too large and complex to perform effectively on such devices, posing challenges for scaling and model deployment. Therefore, striking a compromise between maintaining high accuracy and reducing inference time and model size becomes essential. In the study presented in this white paper, three different models—Custom, VGG16[2], MobileNET[3]—are compressed using tiny machine learning or TinyML, a framework for model optimization and compression. The primary goal is to preserve optimal accuracy while significantly reducing inference time and model size. The study will assess the trade-offs between accuracy, size reduction, and inference time by comparing the compressed models by tailoring and comparing the performance with the original models. Additionally, the study intends to explore TinyML's potential to enhance user experience and enable edge computing in medical applications.

INTRODUCTION:

In recent years, the deployment of deep learning models on resource-constrained edge devices, such as smartphones, wearable devices, IoT devices, edge servers, and embedded systems has increased exponentially, posing challenges due to their limited computational power and memory space. The study presented here, aimed to employ TinyML (tiny machine learning) techniques for compressing Custom, VGG16, and MobileNet models across datasets taken from the fashion, radiology, and dermatology fields. It prioritizes the following three features: achieving optimal accuracy, reduced inference time, and minimized model size suitable for deployment on resource-constrained edge devices. The main compression techniques applied here are quantization and pruning. Quantization makes numbers in the model much less precise, while pruning selectively eliminates redundant or unnecessary connections within the neural network architecture without compromising the model too much. This helps in reducing the computational cost and memory usage for optimal and quick deployment and is adjusted to the requirements of TinyML. For our study, which involved meticulous testing of different permutations and combinations on the datasets, our focus was to tweak the models and parameters to increase accuracy, inference time, and model size. Further, this endeavor leverages quantization techniques from TensorFlow Lite compression to investigate how TinyML might facilitate edge computing and improve user experience. It makes it easier to implement effective and lightweight models,
allowing for real-time inference on edge devices with limited resources.

Access the full report by downloading through the given download button.


That the contents of third-party research report/s published here on the website, and the interpretation of all information in the report/s such as data, maps, numbers etc. displayed in the content and views or the opinions expressed within the content are solely of the author's; and do not reflect the opinions and beliefs of NASSCOM or its affiliates in any manner. NASSCOM does not take any liability w.r.t. content in any manner and will not be liable in any manner whatsoever for any kind of liability arising out of any act, error or omission. The contents of third-party research report/s published, are provided solely as convenience; and the presence of these research report/s should not, under any circumstances, be considered as an endorsement of the contents by NASSCOM in any manner; and if you chose to access these research report/s, you do so at your own risk.


Cyient (Estd: 1991, NSE: CYIENT) delivers Intelligent Engineering solutions across Product, Plant and Network for digital, autonomous and sustainable future to over 300 customers, including 40% of the top 100 global innovators. As a company, Cyient is committed to designing a culturally inclusive, socially responsible, and environmentally sustainable tomorrow together with our stakeholders.



LATEST REPORTS

© Copyright nasscom. All Rights Reserved.