GANchors: Realistic Image Perturbation Distributions for Anchors Using Generative Models
This work incrementally improves trust in model explanations for image classification by producing more realistic perturbations.
The authors tackled the problem of generating realistic image perturbations for anchor-based explanations in image classification by using GANs to sample from a lower-dimensional latent space, resulting in smaller and higher-precision anchors on MNIST and CelebA datasets.
We extend and improve the work of Model Agnostic Anchors for explanations on image classification through the use of generative adversarial networks (GANs). Using GANs, we generate samples from a more realistic perturbation distribution, by optimizing under a lower dimensional latent space. This increases the trust in an explanation, as results now come from images that are more likely to be found in the original training set of a classifier, rather than an overlay of random images. A large drawback to our method is the computational complexity of sampling through optimization; to address this, we implement more efficient algorithms, including a diverse encoder. Lastly, we share results from the MNIST and CelebA datasets, and note that our explanations can lead to smaller and higher precision anchors.