Alessandro De Palma — accepted AI-conference papers

2024

Expressive Losses for Verified Robustness via Convex Combinations

Alessandro De Palma, Rudy R Bunel, Krishnamurthy Dj Dvijotham, M. Pawan Kumar, Robert Stanforth, Alessio Lomuscio

ICLR 2024poster

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 verifiability at the expense of standard performance. As shown in recent work, better trade-offs between accuracy and robu…

Cited by 13 Source ↗PDF ↗code ↗Scholar ↗

2022

In Defense of the Unitary Scalarization for Deep Multi-Task Learning

Vitaly Kurin, Alessandro De Palma, Ilya Kostrikov, Shimon Whiteson, M. Pawan Kumar

NeurIPS 2022accept

Recent multi-task learning research argues against unitary scalarization, where training simply minimizes the sum of the task losses. Several ad-hoc multi-task optimization algorithms have instead been proposed, inspired by various hypotheses about what makes multi-task settings difficult. The majo…

Cited by 85 Source ↗PDF ↗code ↗Scholar ↗

2021

Scaling the Convex Barrier with Active Sets

Alessandro De Palma, Harkirat Behl, Rudy R Bunel, Philip Torr, M. Pawan Kumar

ICLR 2021poster

Tight and efficient neural network bounding is of critical importance for the scaling of neural network verification systems. A number of efficient specialised dual solvers for neural network bounds have been presented recently, but they are often too loose to verify more challenging properties. Thi…

Cited by 71 Source ↗PDF ↗Scholar ↗

2020

Lagrangian Decomposition for Neural Network Verification

Rudy Bunel, Alessandro De Palma, Alban Desmaison, Krishnamurthy Dvijotham, Pushmeet Kohli, Philip Torr +1

UAI 2020poster

A fundamental component of neural network verification is the computation of bounds on the values their outputs can take. Previous methods have either used off-the-shelf solvers, discarding the problem structure, or relaxed the problem even further, making the bounds unnecessarily loose. We propose…

Cited by 65 Source ↗PDF ↗code ↗Scholar ↗