Abstract
We propose to condition a generative model by a given image classifier uncertainty in order to analyze and explain its behavior. Preliminary experiments on synthetic data and a corrupted version of MNIST dataset illustrate the idea.
Explaining an image classifier with a generative model conditioned by uncertainty
Related papers
Ranked by semantic similarity — how closely each paper's abstract matches this one (100% = near-identical topic).
- Cluster-guided Image Synthesis with Unconditional Models (2021)63% match
- Leveraging generative models to characterize the failure conditions of image classifiers (2024)62% match
- Ctrl-U: Robust Conditional Image Generation via Uncertainty-aware Reward Modeling (2024)62% match
- Shape-conditioned Image Generation by Learning Latent Appearance Representation from Unpaired Data (2018)60% match
- Image Decomposition and Classification through a Generative Model (2019)59% match
- UQGAN: A Unified Model for Uncertainty Quantification of Deep Classifiers trained via Conditional GANs (2022)59% match
- Conditional Image Generation by Conditioning Variational Auto-Encoders (2021)59% match
- Leaving Reality to Imagination: Robust Classification via Generated Datasets (2023)59% match