arXiv：cs.LG· Toon Vinck, Na\"in Jonckers, Jaro De Roose, Jeffrey Prinzie, Peter Karsmakers·· 3 小时前AI 评分30
结构化剪枝与容错如何在航天 DNN 中权衡：一项实证研究
Exploring the Trade-Off Between Structured Pruning and Fault Tolerance in Deep Neural Networks for Space Applications
AI 导读
针对航天场景的深度神经网络，研究者通过迭代结构化剪枝缩减模型宽度，并在每个剪枝阶段进行定向故障注入，评估单粒子翻转（SEU）下的表现。结果显示，剪枝虽因冗余减少而提高单次推理的故障敏感度，但更短的执行时间降低了遭遇 SEU 的概率，二者相互抵消。该结论表明结构化剪枝可在不损害整体可靠性的前提下显著节省能耗与延迟。
正文
Abstract:Deep Neural Networks (DNNs) inherently exhibit a degree of robustness to bit-level faults due to their distributed representation of information. As a model increases in width, this information becomes more dispersed, theoretically reducing the impact of any single bit fault. In this paper, we empirically investigate the relationship between model width and robustness to Single Event Upsets (SEUs). We conduct a comprehensive experiment in which baseline models undergo iterative structured pruning to reduce their width while preserving task performance as much as possible. At each pruning stage, we run a targeted fault-injection campaign to evaluate the model's performance under simulated bit-flip scenarios. Our results show that, although structured pruning increases per-inference sensitivity to faults by reducing redundancy, this effect is effectively counterbalanced by shorter execution time, which lowers the probability of encountering an SEU. These findings suggest that structured pruning can yield significant energy and latency savings without compromising overall reliability, providing useful guidance for designing robust AI systems for space applications.
| Comments: | 5 pages, 3 figures, SPAICE 2026 Conference |
| Subjects: | Machine Learning (cs.LG) |
| Cite as: | arXiv:2610.03117 [cs.LG] |
| (or arXiv:2610.03117v1 [cs.LG] for this version) | |
| https://doi.org/10.48550/arXiv.2610.03117 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Toon Vinck [view email]
[v1]
Fri, 2 Oct 2026 10:36:10 UTC (108 KB)
来源：arXiv：cs.LG · arxiv.org