DeltaNN: Assessing the Impact of Computational Environment Parameters on the Performance of Image Recognition Models

TL;DR

DeltaNN generates different implementations of a given image recognition model for variations in environment parameters, namely, deep learning frameworks, compiler optimizations and hardware devices and analyzes differences in model performance as a result and conducts an empirical study of robustness analysis of three popular image recognition models using the ImageNet dataset.

Abstract

Image recognition tasks typically use deep learning and require enormous processing power, thus relying on hardware accelerators like GPUs and TPUs for fast, timely processing. Failure in real-time image recognition tasks can occur due to sub-optimal mapping on hardware accelerators during model deployment, which may lead to timing uncertainty and erroneous behavior. Mapping on hardware accelerators is done using multiple software components like deep learning frameworks, compilers, and device libraries, that we refer to as the computational environment. Owing to the increased use of image recognition tasks in safety-critical applications like autonomous driving and medical imaging, it is imperative to assess their robustness to changes in the computational environment, as the impact of parameters like deep learning frameworks, compiler optimizations, and hardware devices on model performance and correctness is not yet well understood. In this paper we present a differential testing framework, DeltaNN, that allows us to assess the impact of different computational environment parameters on the performance of image recognition models during deployment, post training. DeltaNN generates different implementations of a given image recognition model for variations in environment parameters, namely, deep learning frameworks, compiler optimizations and hardware devices and analyzes differences in model performance as a result. Using DeltaNN, we conduct an empirical study of robustness analysis of three popular image recognition models using the ImageNet dataset. We report the impact in terms of misclassifications and inference time differences across different settings. In total, we observed up to 100% output label differences across deep learning frameworks, and up to 81% unexpected performance degradation in terms of inference time, when applying compiler optimizations.

Figures (10)

• Figure 1: Possible sources of errors when deploying DNN models.
• Figure 2: Differential Testing applied by DeltaNN for a DNN model, varying deep learning frameworks, compiler optimizations, and hardware devices.
• Figure 3: Overview of DNN compilation in Apache TVM.
• Figure 4: Relevant layers in the deep learning systems stack iiswc_2018.
• Figure 5: Architecture of the DeltaNN framework: (1) Model Variant Generator for generating different model implementations when changing and converting DL frameworks and compiler optimizations; (2) Differential Execution for executing the various model implementations on images from a target dataset; and (3) Analysis for comparing output labels and inference time between executions while analyzing source of discrepancy.