Robustar: Interactive Toolbox Supporting Precise Data Annotation for Robust Vision Learning
This work addresses the issue of model robustness for vision learning practitioners by providing an interactive annotation tool, though it is incremental as it builds on existing understanding of spurious features.
The authors tackled the problem of improving robustness in vision classification models by introducing Robustar, a software tool that enables users to annotate spurious features at the pixel level in images, with the goal of removing these features before training to enhance model robustness.
We introduce the initial release of our software Robustar, which aims to improve the robustness of vision classification machine learning models through a data-driven perspective. Building upon the recent understanding that the lack of machine learning model's robustness is the tendency of the model's learning of spurious features, we aim to solve this problem from its root at the data perspective by removing the spurious features from the data before training. In particular, we introduce a software that helps the users to better prepare the data for training image classification models by allowing the users to annotate the spurious features at the pixel level of images. To facilitate this process, our software also leverages recent advances to help identify potential images and pixels worthy of attention and to continue the training with newly annotated data. Our software is hosted at the GitHub Repository https://github.com/HaohanWang/Robustar.