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.

arXiv commentAll authors contributed equally. Appears in AdvML Workshop @ ICML2021: A Blessing in Disguise: The Prospects and Perils of Adversarial Machine Learning

References in corpus (10)

• Adam: A Method for Stochastic Optimization
• Neural Architecture Search with Reinforcement Learning
• Deep Learning for Identifying Metastatic Breast Cancer
• Adversarial Attacks and Defences: A Survey
• Searching for Efficient Multi-Scale Architectures for Dense Image Prediction
• Is feature selection secure against training data poisoning?
• A Survey on Neural Architecture Search
• Evaluating the Search Phase of Neural Architecture Search
• Generative Poisoning Attack Method Against Neural Networks
• Evolving Robust Neural Architectures to Defend from Adversarial Attacks