The Unlikely Hero: Nonideality in Analog Photonic Neural Networks as Built-in Defender Against Adversarial Attacks
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866910628430479360 |
|---|---|
| author | Lu, Haotian Yin, Ziang Bhoumik, Partho Banerjee, Sanmitra Chakrabarty, Krishnendu Gu, Jiaqi |
| author_facet | Lu, Haotian Yin, Ziang Bhoumik, Partho Banerjee, Sanmitra Chakrabarty, Krishnendu Gu, Jiaqi |
| contents | Electronic-photonic computing systems have emerged as a promising platform for accelerating deep neural network (DNN) workloads. Major efforts have been focused on countering hardware non-idealities and boosting efficiency with various hardware/algorithm co-design methods. However, the adversarial robustness of such photonic analog mixed-signal AI hardware remains unexplored. Though the hardware variations can be mitigated with robustness-driven optimization methods, malicious attacks on the hardware show distinct behaviors from noises, which requires a customized protection method tailored to optical analog hardware. In this work, we rethink the role of conventionally undesired non-idealities in photonic analog accelerators and claim their surprising effects on defending against adversarial weight attacks. Inspired by the protection effects from DNN quantization and pruning, we propose a synergistic defense framework tailored for optical analog hardware that proactively protects sensitive weights via pre-attack unary weight encoding and post-attack vulnerability-aware weight locking. Efficiency-reliability trade-offs are formulated as constrained optimization problems and efficiently solved offline without model re-training costs. Extensive evaluation of various DNN benchmarks with a multi-core photonic accelerator shows that our framework maintains near-ideal on-chip inference accuracy under adversarial bit-flip attacks with merely <3% memory overhead. Our codes are open-sourced at https://github.com/ScopeX-ASU/Unlikely_Hero. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2410_01289 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | The Unlikely Hero: Nonideality in Analog Photonic Neural Networks as Built-in Defender Against Adversarial Attacks Lu, Haotian Yin, Ziang Bhoumik, Partho Banerjee, Sanmitra Chakrabarty, Krishnendu Gu, Jiaqi Emerging Technologies Cryptography and Security Optics Electronic-photonic computing systems have emerged as a promising platform for accelerating deep neural network (DNN) workloads. Major efforts have been focused on countering hardware non-idealities and boosting efficiency with various hardware/algorithm co-design methods. However, the adversarial robustness of such photonic analog mixed-signal AI hardware remains unexplored. Though the hardware variations can be mitigated with robustness-driven optimization methods, malicious attacks on the hardware show distinct behaviors from noises, which requires a customized protection method tailored to optical analog hardware. In this work, we rethink the role of conventionally undesired non-idealities in photonic analog accelerators and claim their surprising effects on defending against adversarial weight attacks. Inspired by the protection effects from DNN quantization and pruning, we propose a synergistic defense framework tailored for optical analog hardware that proactively protects sensitive weights via pre-attack unary weight encoding and post-attack vulnerability-aware weight locking. Efficiency-reliability trade-offs are formulated as constrained optimization problems and efficiently solved offline without model re-training costs. Extensive evaluation of various DNN benchmarks with a multi-core photonic accelerator shows that our framework maintains near-ideal on-chip inference accuracy under adversarial bit-flip attacks with merely <3% memory overhead. Our codes are open-sourced at https://github.com/ScopeX-ASU/Unlikely_Hero. |
| title | The Unlikely Hero: Nonideality in Analog Photonic Neural Networks as Built-in Defender Against Adversarial Attacks |
| topic | Emerging Technologies Cryptography and Security Optics |
| url | https://arxiv.org/abs/2410.01289 |