We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:


Current browse context:


Change to browse by:

References & Citations

DBLP - CS Bibliography


(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Computer Science > Machine Learning

Title: Poisoning the Search Space in Neural Architecture Search

Abstract: Deep learning has proven to be a highly effective problem-solving tool for object detection and image segmentation across various domains such as healthcare and autonomous driving. At the heart of this performance lies neural architecture design which relies heavily on domain knowledge and prior experience on the researchers' behalf. More recently, this process of finding the most optimal architectures, given an initial search space of possible operations, was automated by Neural Architecture Search (NAS). In this paper, we evaluate the robustness of one such algorithm known as Efficient NAS (ENAS) against data agnostic poisoning attacks on the original search space with carefully designed ineffective operations. By evaluating algorithm performance on the CIFAR-10 dataset, we empirically demonstrate how our novel search space poisoning (SSP) approach and multiple-instance poisoning attacks exploit design flaws in the ENAS controller to result in inflated prediction error rates for child networks. Our results provide insights into the challenges to surmount in using NAS for more adversarially robust architecture search.
Comments: All authors contributed equally. Appears in AdvML Workshop @ ICML2021: A Blessing in Disguise: The Prospects and Perils of Adversarial Machine Learning
Subjects: Machine Learning (cs.LG); Cryptography and Security (cs.CR); Neural and Evolutionary Computing (cs.NE); Machine Learning (stat.ML)
Cite as: arXiv:2106.14406 [cs.LG]
  (or arXiv:2106.14406v1 [cs.LG] for this version)

Submission history

From: Nayan Saxena [view email]
[v1] Mon, 28 Jun 2021 05:45:57 GMT (2591kb,D)

Link back to: arXiv, form interface, contact.