# Noise-Aware Misclassification Attack Detection in Collaborative DNN Inference

arXiv: https://arxiv.org/abs/2603.17914 · HTML: https://mlsanity.com/paper/2603.17914/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 11 of its 35 references that are in the map.
- First posted on arXiv 18 Mar 2026.

## Builds on (most cited first, 11 of 11)

- [Explaining and Harnessing Adversarial Examples](https://mlsanity.com/paper/1412.6572/) (2014)
- [Evasion Attacks against Machine Learning at Test Time](https://mlsanity.com/paper/1708.06131/) (2017)
- [One pixel attack for fooling deep neural networks](https://mlsanity.com/paper/1710.08864/) (2017)
- [Detecting Adversarial Samples from Artifacts](https://mlsanity.com/paper/1703.00410/) (2017)
- [Dodging Attack Using Carefully Crafted Natural Makeup](https://mlsanity.com/paper/2109.06467/) (2021)
- [adVAE: A self-adversarial variational autoencoder with Gaussian anomaly prior knowledge for anomaly detection](https://mlsanity.com/paper/1903.00904/) (2019)
- [Reinforcement Learning-driven Data-intensive Workflow Scheduling for Volunteer Edge-Cloud](https://mlsanity.com/paper/2407.01428/) (2024)
- [Not All Features Are Equal: Discovering Essential Features for Preserving Prediction Privacy](https://mlsanity.com/paper/2003.12154/) (2020)
- [Infer-EDGE: Dynamic DNN Inference Optimization in 'Just-in-time' Edge-AI Implementations](https://mlsanity.com/paper/2501.18842/) (2025)
- [Detection of Misreporting Attacks on Software-Defined Immersive Environments](https://mlsanity.com/paper/2509.18040/) (2025)
- [MAGE-ID: A Multimodal Generative Framework for Intrusion Detection Systems](https://mlsanity.com/paper/2512.03375/) (2025)

## Abstract (arXiv)

Collaborative inference of object classification Deep neural Networks (DNNs) where resource-constrained end-devices offload partially processed data to remote edge servers to complete end-to-end processing, is becoming a key enabler of edge-AI. However, such edge-offloading is vulnerable to malicious data injections leading to stealthy misclassifications that are tricky to detect, especially in the presence of environmental noise. In this paper, we propose a semi-gray-box and noise- aware anomaly detection framework fueled by a variational autoencoder (VAE) to capture deviations caused by adversarial manipulation. The proposed framework incorporates a robust noise-aware feature that captures the characteristic behavior of environmental noise to improve detection accuracy while reducing false alarm rates. Our evaluation with popular object classification DNNs demonstrate the robustness of the proposed detection (up to 90% AUROC across DNN configurations) under realistic noisy conditions while revealing limitations caused by feature similarity and elevated noise levels.