280 citations · 1k across the 17 of their papers we have counts for
12 papers · 1 filter
Verified Neural Compressed Sensing
Rudy Bunel, Krishnamurthy Dvijotham, M. Pawan Kumar +2
We develop the first (to the best of our knowledge) provably correct neural networks for a precise computational task, with the proof of correctness generated by an automated verif…
Unlocking Accuracy and Fairness in Differentially Private Image Classification
Leonard Berrada, Soham De, Judy Hanwen Shen +6
Privacy-preserving machine learning aims to train models on private data without leaking sensitive information. Differential privacy (DP) is considered the gold standard framework…
Expressive Losses for Verified Robustness via Convex Combinations
Alessandro De Palma, Rudy Bunel, Krishnamurthy Dvijotham +3
In order to train networks for verified adversarial robustness, it is common to over-approximate the worst-case loss over perturbation regions, resulting in networks that attain ve…
Differentially Private Diffusion Models Generate Useful Synthetic Images
Sahra Ghalebikesabi, Leonard Berrada, Sven Gowal +7
The ability to generate privacy-preserving synthetic versions of sensitive image datasets could unlock numerous ML applications currently constrained by data availability. Due to t…
IBP Regularization for Verified Adversarial Robustness via Branch-and-Bound
Alessandro De Palma, Rudy Bunel, Krishnamurthy Dvijotham +2
Recent works have tried to increase the verifiability of adversarially trained networks by running the attacks over domains larger than the original perturbations and adding variou…
Make Sure You're Unsure: A Framework for Verifying Probabilistic Specifications
Leonard Berrada, Sumanth Dathathri, Krishnamurthy Dvijotham +5
Most real world applications require dealing with stochasticity like sensor noise or predictive uncertainty, where formal specifications of desired behavior are inherently probabil…
Contrastive Training for Improved Out-of-Distribution Detection
Jim Winkens, Rudy Bunel, Abhijit Guha Roy +10
Reliable detection of out-of-distribution (OOD) inputs is increasingly understood to be a precondition for deployment of machine learning systems. This paper proposes and investiga…
Are Labels Required for Improving Adversarial Robustness?
Jonathan Uesato, Jean-Baptiste Alayrac, Po-Sen Huang +3
Recent work has uncovered the interesting (and somewhat surprising) finding that training models to be invariant to adversarial perturbations requires substantially larger datasets…
Verification of Non-Linear Specifications for Neural Networks
Chongli Qin, Krishnamurthy, Dvijotham +7
Prior work on neural network verification has focused on specifications that are linear functions of the output of the network, e.g., invariance of the classifier output under adve…
On the Effectiveness of Interval Bound Propagation for Training Verifiably Robust Models
Sven Gowal, Krishnamurthy Dvijotham, Robert Stanforth +6
Recent work has shown that it is possible to train deep neural networks that are provably robust to norm-bounded adversarial perturbations. Most of these methods are based on minim…
Training verified learners with learned verifiers
Krishnamurthy Dvijotham, Sven Gowal, Robert Stanforth +4
This paper proposes a new algorithmic framework, predictor-verifier training, to train neural networks that are verifiable, i.e., networks that provably satisfy some desired input-…
A Dual Approach to Scalable Verification of Deep Networks
Krishnamurthy, Dvijotham, Robert Stanforth +3
This paper addresses the problem of formally verifying desirable properties of neural networks, i.e., obtaining provable guarantees that neural networks satisfy specifications rela…