Exploiting device-level non-idealities for adversarial attacks on ReRAM-based neural networks
Resistive memory (ReRAM) or memristor devices offer the prospect of more efficient computing. While memristors have been used for a variety of computing systems, their usage has gained significant popularity in the domain of deep learning. Weight matrices in deep neural networks can be mapped to cro...
Main Authors: | Tyler McLemore, Robert Sunbury, Seth Brodzik, Zachary Cronin, Elias Timmons, Dwaipayan Chakraborty |
---|---|
Format: | Article |
Language: | English |
Published: |
Elsevier
2023-07-01
|
Series: | Memories - Materials, Devices, Circuits and Systems |
Subjects: | |
Online Access: | http://www.sciencedirect.com/science/article/pii/S2773064623000300 |
Similar Items
-
Crossbar-constrained technology mapping for ReRAM based in-memory computing
by: Bhattacharjee, Debjyoti, et al.
Published: (2021) -
Effective Current-Driven Memory Operations for Low-Power ReRAM Applications
by: Albert Cirera, et al.
Published: (2023-01-01) -
Digital ReRAM-based compute-in-memory design
by: Xu, Jiawei
Published: (2024) -
Thermal Heating in ReRAM Crossbar Arrays: Challenges and Solutions
by: Kamilya Smagulova, et al.
Published: (2024-01-01) -
Utilizing the Switching Stochasticity of HfO2/TiOx-Based ReRAM Devices and the Concept of Multiple Device Synapses for the Classification of Overlapping and Noisy Patterns
by: Christopher Bengel, et al.
Published: (2021-06-01)