CheckOne: Lightweight Fault Detection and Mitigation for Vision Transformers
Mohammad Hasan Ahmadilivani, Sven-Markus Loorits, Jaan Raik
Abstract
The wide adoption of Vision Transformers (ViTs) in safety-critical applications raises reliability concerns related to hardware faults. Algorithm-Based Fault Tolerance (ABFT) methods have emerged as lightweight and symmetric protection mechanisms for DNNs. However, they are particularly challenging for ViTs due to their significant computational requirements. This work comprehensively evaluates the reliability of ViTs, emphasizing the need for symmetric protection in their layers. Furthermore, we present CheckOne, a novel, cost-effective method for fault detection and mitigation in ViTs that significantly reduces the computational cost compared to conventional ABFT. Through extensive experiments with multiple ViTs, CheckOne mitigates critical faults by up to 26× and achieves an average 3.8x higher performance than ABFT in ViTs.
Create a lesson
Related papers
Locus: A Framework for Exploring and Optimizing Point Addition Hardware for Zero-Knowledge Proofs
Gaurav Kuwar, Alhad Daftardar, Jianqiao Mo et al.
Quantifying the Effect of HCLs on a Fixed-Microarchitecture MXFP4 Accelerator
Daniele Passaretti, Sajjad Tamimi, Nicola Dall'Ora
HBFlex: A Flexible Memory System for Bridging Fine-Grained LLM States and Coarse-Grained HBF Parallel Execution
Shuzhang Zhong, Weikai Xu, Yifan Zhou et al.
Automated Instruction Encoding Synthesis for Modern GPU ISA Compression
Mingyuan Ma, Hu He
VeriBugBench: An Empirically Grounded Framework for Constructing Verilog RTL Debugging Benchmarks
Xiankai Meng, Kejian Feng, Xinlin Zhao et al.
Budgeted Express-Mesh: Traffic-Aware Link Placement and Deadlock-Free Adaptive Routing
Li Cao, Jingyuan Ma