Search

Pushmeet Kohli

A framework for robustness certification of smoothed classifiers using f-divergences
Achieving robustness in the wild via adversarial mixing with disentangled representations
Efficient neural network verification with exactness characterization
Lagrangian decomposition for neural network verification
The Autoencoding Variational Autoencoder
Towards Verified Robustness under Text Deletion Interventions
A Dual Approach to Verify and Train Deep Networks.
Achieving verified robustness to symbol substitutions via interval bound propagation
Adversarial robustness through local linearization
Adversarially robust representations with smooth encoders
Knowing when to stop: Evaluation and verification of conformity to output-size specifications
Scalable verified training for provably robust image classification
Toward evaluating robustness of deep reinforcement learning with continuous control
Verification of non-linear specifications for neural networks
A dual approach to scalable verification of deep networks

Published with Wowchemy — the free, open source website builder that empowers creators.