Timezone: »

 
Spotlight
Scalable Certified Segmentation via Randomized Smoothing
Marc Fischer · Maximilian Baader · Martin Vechev

Thu Jul 22 06:45 AM -- 06:50 AM (PDT) @
in Adversarial Learning 1 »

We present a new certification method for image and point cloud segmentation based on randomized smoothing. The method leverages a novel scalable algorithm for prediction and certification that correctly accounts for multiple testing, necessary for ensuring statistical guarantees. The key to our approach is reliance on established multiple-testing correction mechanisms as well as the ability to abstain from classifying single pixels or points while still robustly segmenting the overall input. Our experimental evaluation on synthetic data and challenging datasets, such as Pascal Context, Cityscapes, and ShapeNet, shows that our algorithm can achieve, for the first time, competitive accuracy and certification guarantees on real-world segmentation tasks. We provide an implementation at https://github.com/eth-sri/segmentation-smoothing.

Keywords: Social Aspects of Machine Learning · AI Safety

Author Information

Marc Fischer (ETH Zurich)
Maximilian Baader (ETH Zürich)
Martin Vechev (ETH Zurich)

Related Events (a corresponding poster, oral, or spotlight)

More from the Same Authors