Cache Telepathy: Leveraging Shared Resource Attacks to Learn DNN Architectures

2018-08-14 · Mengjia Yan, Christopher Fletcher, Josep Torrellas

Deep Neural Networks (DNNs) are fast becoming ubiquitous for their ability to attain good accuracy in various machine learning tasks. A DNN's architecture (i.e., its hyper-parameters) broadly determines the DNN's accuracy and performance, and is often confidential. Attacking a DNN in the cloud to obtain its architecture can potentially provide major commercial value. Further, attaining a DNN's architecture facilitates other, existing DNN attacks. This paper presents Cache Telepathy: a fast and accurate mechanism to steal a DNN's architecture using the cache side channel. Our attack is based on the insight that DNN inference relies heavily on tiled GEMM (Generalized Matrix Multiply), and that DNN architecture parameters determine the number of GEMM calls and the dimensions of the matrices used in the GEMM functions. Such information can be leaked through the cache side channel. This paper uses Prime+Probe and Flush+Reload to attack VGG and ResNet DNNs running OpenBLAS and Intel MKL libraries. Our attack is effective in helping obtain the architectures by very substantially reducing the search space of target DNN architectures. For example, for VGG using OpenBLAS, it reduces the search space from more than $10^{35}$ architectures to just 16.

📄 PDF Abstract BibTeX arXiv:1808.04761

Code (0)

등록된 구현이 없습니다.

Methods 이 논문이 사용한 방법론

Average Pooling 설명 없음

Dropout Dropout is a regularization technique for neural networks that drops a unit (along with connections) at training time with a specified probability $p$ (a common value is…

1x1 Convolution A 1 x 1 Convolution is a convolution with some special properties in that it can be used for dimensionality reduction,…

Batch Normalization 설명 없음

Bottleneck Residual Block A Bottleneck Residual Block is a variant of the residual block that utilises 1x1 convolutions to create a bottleneck. The…

Global Average Pooling Global Average Pooling is a pooling operation designed to replace fully connected layers in classical CNNs. The idea is to generate one feature map for each corresponding…

Residual Block Residual Blocks are skip-connection blocks that learn residual functions with reference to the layer inputs, instead of learning unreferenced functions. They were introduced…

Dense Connections Dense Connections, or Fully Connected Connections, are a type of layer in a deep neural network that use a linear operation where every input is connected to every output…

Hybrid Deep Learning Model for Multiple Cache Side Channel Attacks Detection: A Comparative Analysis

2025-01-28 · Tejal Joshi, Aarya Kawalay, Anvi Jamkhande, Amit Joshi

Cache side channel attacks are a sophisticated and persistent threat that exploit vulnerabilities in modern processors to extract sensitive information. These attacks leverage weaknesses in shared computational resources…

CacheProbe: Auditing Prompt Cache Isolation in Gateway APIs

2026-05-28 · Ryan Fahey arxiv

Over the past year, prompt caching in Large Language Models (LLMs) has become increasingly more popular across inference APIs. Prompt caching helps save precious compute resources and speeds up response times by reusing …

Spill The Beans: Exploiting CPU Cache Side-Channels to Leak Tokens from Large Language Models

2025-05-01 · Andrew Adiletta, Berk Sunar

Side-channel attacks on shared hardware resources increasingly threaten confidentiality, especially with the rise of Large Language Models (LLMs). In this work, we introduce Spill The Beans, a novel application of cache …

CPU

Accelerating Suffix Jailbreak attacks with Prefix-Shared KV-cache

2026-03-12 · Xinhai Wang, Shaopeng Fu, Shu Yang, Liangyu Wang 외 arxiv

Suffix jailbreak attacks serve as a systematic method for red-teaming Large Language Models (LLMs) but suffer from prohibitive computational costs, as a large number of candidate suffixes need to be evaluated before iden…

Attacking Slicing Network via Side-channel Reinforcement Learning Attack

2024-09-17 · Wei Shao, Chandra Thapa, Rayne Holland, Sarah Ali Siddiqui 외

Network slicing in 5G and the future 6G networks will enable the creation of multiple virtualized networks on a shared physical infrastructure. This innovative approach enables the provision of tailored networks to accom…

reinforcement-learning Reinforcement Learning