验证

训练稳健网络并验证网络稳健性

深度学习验证是一套评估深度神经网络属性的方法。例如，您可以验证网络的稳健性属性、计算网络输出边界、查找对抗示例以及检测分布外数据，并验证与行业标准的合规性。

Deep Learning Toolbox Verification Library 支持包能够测试深度学习网络的稳健性属性。

使用 verifyNetworkRobustness 函数验证网络对对抗示例的稳健性。当在指定的输入下界和上界之间扰动输入时，如果网络的预测类不改变，则表明网络针对对抗输入具有稳健性。对于一组输入边界，该函数检查网络对于那些输入边界之间的对抗示例是否具有稳健性，并返回 verified、violated 或 unproven。
当输入介于指定的下界和上界之间时，使用 estimateNetworkOutputBounds 函数来估计网络返回的输出值的范围。使用此函数估计网络预测对输入扰动的敏感度。
使用 networkDistributionDiscriminator 函数创建一个分布判别器，该判别器将数据分为分布内和分布外。

函数

`estimateNetworkOutputBounds`	Estimate output bounds of deep learning network (自 R2022b 起)
`verifyNetworkRobustness`	Verify adversarial robustness of deep learning network (自 R2022b 起)
`networkDistributionDiscriminator`	Deep learning distribution discriminator (自 R2023a 起)
`isInNetworkDistribution`	Determine whether data is within the distribution of the network (自 R2023a 起)
`distributionScores`	Distribution confidence scores (自 R2023a 起)

对象

`BaselineDistributionDiscriminator`	Baseline distribution discriminator (自 R2023a 起)
`EnergyDistributionDiscriminator`	Energy distribution discriminator (自 R2023a 起)
`ODINDistributionDiscriminator`	ODIN distribution discriminator (自 R2023a 起)
`HBOSDistributionDiscriminator`	HBOS distribution discriminator (自 R2023a 起)

主题

Verify Robustness of Deep Learning Neural Network
This example shows how to verify the adversarial robustness of a deep learning neural network.
Deep Learning Visualization Methods
Learn about and compare deep learning visualization methods.
Out-of-Distribution Detection for Deep Neural Networks
This example shows how to detect out-of-distribution (OOD) data in deep neural networks.
Verify an Airborne Deep Learning System
This example shows how to verify a deep learning system for airborne applications and is based on the work in [5,6,7], which includes the development and verification activities required by DO-178C [1], ARP4754A [2], and prospective EASA and FAA guidelines [3,4]. (自 R2023b 起)

精选示例

Generate Untargeted and Targeted Adversarial Examples for Image Classification

Use the fast gradient sign method (FGSM) and the basic iterative method (BIM) to generate adversarial examples for a pretrained neural network.

打开实时脚本

Train Image Classification Network Robust to Adversarial Examples

Train a neural network that is robust to adversarial examples using fast gradient sign method (FGSM) adversarial training.

打开实时脚本

Train Robust Deep Learning Network with Jacobian Regularization

Train a neural network that is robust to adversarial examples using a Jacobian regularization scheme.

打开实时脚本

Generate Adversarial Examples for Semantic Segmentation

Generate adversarial examples for a semantic segmentation network using the basic iterative method (BIM).

打开实时脚本

Explore Semantic Segmentation Network Using Grad-CAM

Explore the predictions of a pretrained semantic segmentation network using Grad-CAM.

打开实时脚本

Interpret Deep Network Predictions on Tabular Data Using LIME

Use the locally interpretable model-agnostic explanations (LIME) technique to understand the predictions of a deep neural network classifying tabular data. You can use the LIME technique to understand which predictors are most important to the classification decision of a network.

打开实时脚本