Special Session: Towards an Agile Design Methodology for Efficient, Reliable, and Secure ML Systems

Dave, Shail; Marchisio, Alberto; Hanif, Muhammad Abdullah; Guesmi, Amira; Shrivastava, Aviral; Alouani, Ihsen; Shafique, Muhammad

doi:10.1109/vts52500.2021.9794253

Cited by 11 publications

(3 citation statements)

References 127 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For these reasons, testability and dependability of AI hardware accelerators are important issues that need to be addressed already from the design phase [12]. Inspiration can of course be drawn from known and mature methodologies applied to traditional computer architectures, but the architectural particularities of AI hardware accelerators often make such methodologies prohibitive in terms of cost and quality, requiring the development of new methodologies that are better suited and take full advantage of the said architectural particularities.…”

Section: Introductionmentioning

confidence: 99%

Testability and Dependability of AI Hardware: Survey, Trends, Challenges, and Perspectives

Liu

Stratigopoulos³

2023

IEEE Des. Test

View full text Add to dashboard Cite

In recent years, there has been an expedited trend in embracing bold and radical innovation of computer architectures, aiming at the continuation of computing performance improvement despite the slowed-down physical device scaling. One new frontier in this field focuses on Artificial Intelligence (AI) hardware. While functionality of AI hardware still remains the main focus, testability and dependability of these new architectures need to be addressed before the mainstream adoption. This survey paper covers the state-of-the-art in research and development of dependability and testability solutions for AI hardware including digital or analog implementations of Artificial Neural Networks (ANNs) and Spiking Neural Networks (SNNs), used in accelerators and neuromorphic designs. Trends, challenges and perspectives are also discussed in this paper.

show abstract

Section: Introductionmentioning

confidence: 99%

Testability and Dependability of AI Hardware: Survey, Trends, Challenges, and Perspectives

Liu

Stratigopoulos³

2023

IEEE Des. Test

View full text Add to dashboard Cite

show abstract

“…Machine learning (ML) has witnessed significant development in recent years, finding diverse applications in various sectors such as robotics, automotive, smart industries, economics, medicine, and security [1][2][3]. Several models based on the structure of the human brain have been implemented [4], including the widely used deep neural networks (DNNs) [5,6] and spiking neural networks (SNNs) [7], which emulate the functioning of neurons relatively better than DNNs [8].…”

Section: Introductionmentioning

confidence: 99%

A Homomorphic Encryption Framework for Privacy-Preserving Spiking Neural Networks

Nikfam,

Casaburi,

Marchisio

et al. 2023

Information

Self Cite

View full text Add to dashboard Cite

Machine learning (ML) is widely used today, especially through deep neural networks (DNNs); however, increasing computational load and resource requirements have led to cloud-based solutions. To address this problem, a new generation of networks has emerged called spiking neural networks (SNNs), which mimic the behavior of the human brain to improve efficiency and reduce energy consumption. These networks often process large amounts of sensitive information, such as confidential data, and thus privacy issues arise. Homomorphic encryption (HE) offers a solution, allowing calculations to be performed on encrypted data without decrypting them. This research compares traditional DNNs and SNNs using the Brakerski/Fan-Vercauteren (BFV) encryption scheme. The LeNet-5 and AlexNet models, widely-used convolutional architectures, are used for both DNN and SNN models based on their respective architectures, and the networks are trained and compared using the FashionMNIST dataset. The results show that SNNs using HE achieve up to 40% higher accuracy than DNNs for low values of the plaintext modulus t, although their execution time is longer due to their time-coding nature with multiple time steps.

show abstract

“…Owners of data covertly transfer their information to servers that train a certain neural network and break the law. SecureML trains a neural network with secure account protocols utilizing a more effective custom activation function [46], [47]. The managed model is afterwards secretly distributed across the servers at the conclusion of the account.…”

mentioning

confidence: 99%

A review of Deep Learning Privacy, Security and Defenses

Dawood,

Hadi

2023

Technium

View full text Add to dashboard Cite

Deep learning (DL) can be considered as a powerful tool in different fields and for different applications but its importance raised the concern about privacy, security, and defense issues. This research presents an important overview about different aspects and state-of-the-art techniques in DL privacy, security, and defense. Wide range of topics was covered including private data frameworks, different types of threats and attacks, and the most important defense techniques. We have also discussed the challenges and limitations of each approach besides to possible future research directions. This survey can be considered as a comprehensive guide for other researchers and policymakers who are interested in understanding these important topics associated with DL.

show abstract

Special Session: Towards an Agile Design Methodology for Efficient, Reliable, and Secure ML Systems

Cited by 11 publications

References 127 publications

Testability and Dependability of AI Hardware: Survey, Trends, Challenges, and Perspectives

Testability and Dependability of AI Hardware: Survey, Trends, Challenges, and Perspectives

A Homomorphic Encryption Framework for Privacy-Preserving Spiking Neural Networks

A review of Deep Learning Privacy, Security and Defenses

Contact Info

Product

Resources

About