PEAT: Pseudo-Error Assessment for GPU Kernel Validation in DNN Training

• URL: http://arxiv.org/abs/2609.13544v1
• Date: Fri, 11 Sep 2026 21:16:15 GMT
• Title: PEAT: Pseudo-Error Assessment for GPU Kernel Validation in DNN Training
• Abstract summary: We present PEAT - a lightweight inspection framework for underlinePseudo-underlineError underlineAssessment associated with GPU kernel validation in DNN underlineTraining.<n>Inspired by conventional fault injection (FI), PEAT's Profiler invokes an operation-wise kernel in a training flow to collect a model's states.<n>PEAT's Analyzer characterizes profiled errors, revealing some signatures from the error distribution of a kernel compared to the golden one.
• Score: 1.0132958091492892
• License: http://creativecommons.org/licenses/by-nc-sa/4.0/
• Abstract: Deep neural networks (DNNs) are widely adopted in various fields, driving an emerging trend in developing software stacks associated with DNN training systems. For example, many codes have been ported across different frameworks or developed to leverage the computing power of GPUs or domain-specific accelerators. However, validating a kernel implementation in DNN training is time-consuming and generally requires massive storage. Specifically, this poses a fundamental question: how to characterize the behavior of a new implementation when it is integrated into a DNN training flow. Unfortunately, this problem is not well investigated in the literature, to the best of our knowledge. To address this shortcoming, we present PEAT - a lightweight inspection framework for \underline{P}seudo-\underline{E}rror \underline{A}ssessment associated with GPU kernel validation in DNN \underline{T}raining. Firstly, inspired by conventional fault injection (FI), PEAT's Profiler invokes an operation-wise kernel in a training flow to collect a DNN model's states (e.g., checkpoints and activations). More importantly, the Profiler introduces two simple yet effective techniques, playback FI and frequency-based runtime FI, leveraging persistent kernel calling during the training process. Secondly, PEAT's Analyzer characterizes profiled errors, revealing some signatures from the error distribution of a kernel compared to the golden one. Lastly, PEAT's Detector provides some guidelines as a sufficient condition, which enables associating several well-known error models with signature patterns. We demonstrate the applicability of our approach by presenting the results and analysis using GPUs from the two most popular vendors, NVIDIA V100 and AMD MI250, on various AI models, from vision tasks to language models, for both pretraining and finetuning scenarios.

Related papers

• Achieving Near-Zero-Overhead Multi-Model Hierarchical Classification in Real-Time Detection Pipelines [0.0]
  Edge-deployed vision systems in target recognition, surveillance, autonomous vehicles, and drone domains require hierarchical inference pipelines.<n>Running all models on the GPU creates a serial bottleneck that limits real-time throughput as pipeline stages grow.<n>We present a five-step methodology for zero fallback DLA backbones.
  arXiv  Detail & Related papers  (2026-08-12T08:14:22Z)
• ChipMATE: Multi-Agent Training via Reinforcement Learning for Enhanced RTL Generation [55.947962672433675]
  ChipMATE is the first self-trained multi-agent framework for RTL generation.<n>Inspired by industrial practice, ChipMATE pairs a Verilog agent with a Python reference-model agent that mutually verify each other's outputs.<n>ChipMATE achieves 75.0% and 80.1% pass@1 on VerilogEval V2 with 4B and 9B base models.
  arXiv  Detail & Related papers  (2026-05-13T01:04:21Z)
• Speeding Up Mixed-Integer Programming Solvers with Sparse Learning for Branching [0.5352699766206809]
  We focus on approximating strong branching (SB) scores, a highly effective yet computationally expensive branching rule.<n>We build models with fewer than 4% of the parameters of a state-of-the-art graph neural network (GNN)<n>Relative to SCIP's built-in branching rules and the GNN-based model, our CPU-only models are faster than the default solver and the GPU-accelerated GNN.
  arXiv  Detail & Related papers  (2026-03-31T18:17:56Z)
• Generative Modeling of Regular and Irregular Time Series Data via Koopman VAEs [50.25683648762602]
  We introduce Koopman VAE, a new generative framework that is based on a novel design for the model prior. Inspired by Koopman theory, we represent the latent conditional prior dynamics using a linear map. KoVAE outperforms state-of-the-art GAN and VAE methods across several challenging synthetic and real-world time series generation benchmarks.
  arXiv  Detail & Related papers  (2023-10-04T07:14:43Z)
• Exploring Tradeoffs in Spiking Neural Networks [0.0]
  Spiking Neural Networks (SNNs) have emerged as a promising alternative to traditional Deep Neural Networks for low-power computing. We show that relaxing the spike constraint provides higher performance while also benefiting from faster convergence, similar sparsity, comparable prediction latency, and better robustness to noise compared to TTFS SNNs.
  arXiv  Detail & Related papers  (2022-12-15T20:18:14Z)
• Intelligence Processing Units Accelerate Neuromorphic Learning [52.952192990802345]
  Spiking neural networks (SNNs) have achieved orders of magnitude improvement in terms of energy consumption and latency. We present an IPU-optimized release of our custom SNN Python package, snnTorch.
  arXiv  Detail & Related papers  (2022-11-19T15:44:08Z)
• Testing Feedforward Neural Networks Training Programs [13.249453757295083]
  Multiple testing techniques are proposed to generate test cases that can expose inconsistencies in the behavior of Deep Neural Networks. These techniques assume implicitly that the training program is bug-free and appropriately configured. We propose TheDeepChecker, an end-to-end property-based debug approach for DNN training programs.
  arXiv  Detail & Related papers  (2022-04-01T20:49:14Z)
• DNN Training Acceleration via Exploring GPGPU Friendly Sparsity [16.406482603838157]
  We propose the Approximate Random Dropout that replaces the conventional random dropout of neurons and synapses with a regular and online generated row-based or tile-based dropout patterns. We then develop a SGD-based Search Algorithm that produces the distribution of row-based or tile-based dropout patterns to compensate for the potential accuracy loss. We also propose the sensitivity-aware dropout method to dynamically drop the input feature maps based on their sensitivity so as to achieve greater forward and backward training acceleration.
  arXiv  Detail & Related papers  (2022-03-11T01:32:03Z)
• Rethinking Nearest Neighbors for Visual Classification [56.00783095670361]
  k-NN is a lazy learning method that aggregates the distance between the test image and top-k neighbors in a training set. We adopt k-NN with pre-trained visual representations produced by either supervised or self-supervised methods in two steps. Via extensive experiments on a wide range of classification tasks, our study reveals the generality and flexibility of k-NN integration.
  arXiv  Detail & Related papers  (2021-12-15T20:15:01Z)
• Online Limited Memory Neural-Linear Bandits with Likelihood Matching [53.18698496031658]
  We study neural-linear bandits for solving problems where both exploration and representation learning play an important role. We propose a likelihood matching algorithm that is resilient to catastrophic forgetting and is completely online.
  arXiv  Detail & Related papers  (2021-02-07T14:19:07Z)
• Binary Graph Neural Networks [69.51765073772226]
  Graph Neural Networks (GNNs) have emerged as a powerful and flexible framework for representation learning on irregular data. In this paper, we present and evaluate different strategies for the binarization of graph neural networks. We show that through careful design of the models, and control of the training process, binary graph neural networks can be trained at only a moderate cost in accuracy on challenging benchmarks.
  arXiv  Detail & Related papers  (2020-12-31T18:48:58Z)
• Learning Low-rank Deep Neural Networks via Singular Vector Orthogonality Regularization and Singular Value Sparsification [53.50708351813565]
  We propose SVD training, the first method to explicitly achieve low-rank DNNs during training without applying SVD on every step. We empirically show that SVD training can significantly reduce the rank of DNN layers and achieve higher reduction on computation load under the same accuracy.
  arXiv  Detail & Related papers  (2020-04-20T02:40:43Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.