Nonideality-aware training makes memristive networks more robust to adversarial attacks
Open Access
- 11 February 2025
- journal article
- Published by AIP Publishing in APL Machine Learning
- Vol. 3 (1)
Abstract
Neural networks are now deployed in a wide number of areas, from object classification to natural language systems. Implementations using analog devices such as memristors promise better power efficiency, potentially bringing these applications to a greater number of environments. However, such systems suffer from more frequent device faults, and overall, their exposure to adversarial attacks has not been studied extensively. In this work, we investigate how nonideality-aware training—a common technique to deal with physical nonidealities—affects adversarial robustness. We find that adversarial robustness is significantly improved, even with limited knowledge of what nonidealities will be encountered during test time.Keywords
All Related Versions
Funding Information
- Royal Academy of Engineering (ICRF2223-6-158)
- EPSRC (EP/X018431/1)
This publication has 10 references indexed in Scilit:
- Emerging Nonvolatile Memories for Machine Learning2024
- Nonideality‐Aware Training for Accurate and Robust Low‐Power Memristive Neural NetworksAdvanced Science, 2022
- 4K-memristor analog-grade passive crossbar circuitNature Communications, 2021
- Accurate deep neural network inference using computational phase-change memoryNature Communications, 2020
- Fully hardware-implemented memristor convolutional neural networkNature, 2020
- A survey of deep learning techniques for autonomous drivingJournal of Field Robotics, 2019
- Simulation of Inference Accuracy Using Realistic RRAM DevicesFrontiers in Neuroscience, 2019
- Neuro-Inspired Computing With Emerging Nonvolatile MemorysProceedings of the IEEE, 2018
- Adversarial machine learning2011
- Can machine learning be secure?2006