Survival of the Cheapest: Cost-Aware Hardware Adaptation for Adversarial Robustness

TL;DR

This paper introduces a cost-aware framework for adaptive hardware and hyper-parameter selection in adversarially robust machine learning, optimizing robustness, cost, and latency in cloud environments.

Contribution

It presents a novel decision-support framework using AFT models for dynamic hardware and hyper-parameter adaptation to improve robustness and cost-efficiency.

Findings

Nvidia L4 increases adversarial survival time by 20% at 75% lower cost than V100.

Hardware cost does not necessarily correlate with increased robustness.

Inference latency is a stronger predictor of robustness than training time or hardware.

Abstract

Deploying adversarially robust machine learning systems requires continuous trade-offs between robustness, cost, and latency. We present an autonomic decision-support framework providing a quantitative foundation for adaptive hardware selection and hyper-parameter tuning in cloud-native deep learning. The framework applies accelerated failure time (AFT) models to quantify the effect of hardware choice, batch size, epochs, and validation accuracy on model survival time. This framework can be naturally integrated into an autonomic control loop (monitor--analyse--plan--execute, MAPE-K), where system metrics such as cost, robustness, and latency are continuously evaluated and used to adapt model configurations and hardware selection. Experiments across three GPU architectures confirm the framework is both sound and cost-effective: the Nvidia L4 yields a 20% increase in adversarial survival time while costing 75% less than the V100, demonstrating that expensive hardware does not necessarily improve robustness. The analysis further reveals that model inference latency is a stronger predictor of adversarial robustness than training time or hardware configuration.