Exploiting Efficiency Vulnerabilities in Dynamic Deep Learning Systems

TL;DR

This paper investigates security vulnerabilities in dynamic deep learning systems where input-dependent computation can be exploited by adversaries to cause excessive latency and resource consumption, highlighting the need for robust defenses.

Contribution

It provides a survey of existing attack strategies on DDLSs, identifies gaps in current defenses, and proposes to develop targeted methods to defend against efficiency-based adversarial attacks.

Findings

Dynamic behaviors in DDLSs can be exploited to induce high latency and energy consumption.

Current defenses are insufficient against input-dependent efficiency attacks.

The paper outlines a plan to evaluate attack feasibility and develop robust countermeasures.

Abstract

The growing deployment of deep learning models in real-world environments has intensified the need for efficient inference under strict latency and resource constraints. To meet these demands, dynamic deep learning systems (DDLSs) have emerged, offering input-adaptive computation to optimize runtime efficiency. While these systems succeed in reducing cost, their dynamic nature introduces subtle and underexplored security risks. In particular, input-dependent execution pathways create opportunities for adversaries to degrade efficiency, resulting in excessive latency, energy usage, and potential denial-of-service in time-sensitive deployments. This work investigates the security implications of dynamic behaviors in DDLSs and reveals how current systems expose efficiency vulnerabilities exploitable by adversarial inputs. Through a survey of existing attack strategies, we identify gaps in the coverage of emerging model architectures and limitations in current defense mechanisms. Building on these insights, we propose to examine the feasibility of efficiency attacks on modern DDLSs and develop targeted defenses to preserve robustness under adversarial conditions.