mlsanityData to 14 Oct 2026

Efficient Soft-Error Detection for Low-precision Deep Learning Recommendation Models

arXiv 2103.00130 ↗First posted 27 Feb 2021cs.DCcited by 4

Not on the map. Why?

Abstract (arXiv)

Soft error, namely silent corruption of signal or datum in a computer system, cannot be caverlierly ignored as compute and communication density grow exponentially. Soft error detection has been studied in the context of enterprise computing, high-performance computing and more recently in convolutional neural networks related to autonomous driving. Deep learning recommendation systems (DLRMs) have by now become ubiquitous and serve billions of users per day. Nevertheless, DLRM-specific soft error detection methods are hitherto missing. To fill the gap, this paper presents the first set of soft-error detection methods for low-precision quantized-arithmetic operators in DLRM including general matrix multiplication (GEMM) and EmbeddingBag. A practical method must detect error and do so with low overhead lest reduced inference speed degrades user experience. Exploiting the characteristics of both quantized arithmetic and the operators, we achieved more than 95% detection accuracy for GEMM with an overhead below 20%. For EmbeddingBag, we achieved 99% effectiveness in significant-bit-flips detection with less than 10% of false positives, while keeping overhead below 26%.

Key facts. Cited by 4 papers in the map (data through 14 Oct 2026). Builds on 8 of its 31 references that are in the map. First posted on arXiv 27 Feb 2021.

Builds on

8 of 31 references.

References · 8

  1. 2017Quantization and Training of Neural Networks for Efficient Integer-Arithmetic-Only Inferencecited by 1,593
  2. 2020Distributed Hierarchical GPU Parameter Server for Massive Scale Deep Learning Ads Systemscited by 60
  3. 2021FBGEMM: Enabling High-Performance Low-Precision Deep Learning Inferencecited by 41
  4. 2021Silent Data Corruptions at Scalecited by 23
  5. 2019Post-Training 4-bit Quantization on Embedding Tablescited by 22
  6. 2018ThUnderVolt: Enabling Aggressive Voltage Underscaling and Timing Error Resilience for Energy Efficient Deep Neural Network Acceleratorscited by 18
  7. 2020Making Convolutions Resilient via Algorithm-Based Error Detection Techniquescited by 12
  8. 2018Improving Performance of Iterative Methods by Lossy Checkpontingcited by 3

Built on by

Cited by year

Citing papers per year: 2021: 1; 2022: 0; 2023: 2; 2024: 1.

4 papers.

  1. 2024ATTNChecker: Highly-Optimized Fault Tolerant Attention for Large Language Model Trainingcited by 12
  2. 2021Arithmetic-Intensity-Guided Fault Tolerance for Neural Network Inference on GPUscited by 9
  3. 2023Evaluating and Enhancing Robustness of Deep Recommendation Systems Against Hardware Errorscited by 1
  4. 2023Exploring Winograd Convolution for Cost-effective Neural Network Fault Tolerancecited by 1

Similar, not cited

Close in wording, no citation either way

  1. 2024Memory Faults in Activation-sparse Quantized Deep Neural Networks: Analysis and Mitigation using Sharpness-aware Training
  2. 2026StatQAT: Statistical Quantizer Optimization for Deep Networks
  3. 2024DQRM: Deep Quantized Recommendation Models
  4. 2025MSQ: Memory-Efficient Bit Sparsification Quantization
  5. 2026MCEL: Margin-Based Cross-Entropy Loss for Error-Tolerant Quantized Neural Networks
  6. 2025Low-bit Model Quantization for Deep Neural Networks: A Survey
  7. 2026V-ABFT: Variance-Based Adaptive Threshold for Fault-Tolerant Matrix Multiplication in Mixed-Precision Deep Learning
  8. 2022DenseShift: Towards Accurate and Efficient Low-Bit Power-of-Two Quantization