ONNXExplainer: an ONNX Based Generic Framework to Explain Neural Networks Using Shapley Values
This addresses the need for cross-platform, efficient explanation tools for practitioners using neural networks, though it is incremental as it builds on existing Shapley value methods.
The paper tackles the problem of inefficient and platform-specific neural network explanation methods by introducing ONNXExplainer, a generic framework using Shapley values in the ONNX ecosystem, which improves explanation latency by up to 500% for models like VGG19 and ResNet50.
Understanding why a neural network model makes certain decisions can be as important as the inference performance. Various methods have been proposed to help practitioners explain the prediction of a neural network model, of which Shapley values are most popular. SHAP package is a leading implementation of Shapley values to explain neural networks implemented in TensorFlow or PyTorch but lacks cross-platform support, one-shot deployment and is highly inefficient. To address these problems, we present the ONNXExplainer, which is a generic framework to explain neural networks using Shapley values in the ONNX ecosystem. In ONNXExplainer, we develop its own automatic differentiation and optimization approach, which not only enables One-Shot Deployment of neural networks inference and explanations, but also significantly improves the efficiency to compute explanation with less memory consumption. For fair comparison purposes, we also implement the same optimization in TensorFlow and PyTorch and measure its performance against the current state of the art open-source counterpart, SHAP. Extensive benchmarks demonstrate that the proposed optimization approach improves the explanation latency of VGG19, ResNet50, DenseNet201, and EfficientNetB0 by as much as 500%.