Scaling the convex barrier with active sets

Conference item

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. This lack of tightness is linked to the weakness of the employed relaxation, which is usually a linear program of size linear in the number of neurons. While a tighter linear relaxation for piecewise linear activations exists, it comes at the cost of exponentially many constraints and thus currently lacks an efficient customised solver. We alleviate this deficiency via a novel dual algorithm that realises the full potential of the new relaxation by operating on a small active set of dual variables. Our method recovers the strengths of the new relaxation in the dual space: tightness and a linear separation oracle. At the same time, it shares the benefits of previous dual approaches for weaker relaxations: massive parallelism, GPU implementation, low cost per iteration and valid bounds at any time. As a consequence, we obtain better bounds than off-the-shelf solvers in only a fraction of their running time and recover the speed-accuracy trade-offs of looser dual solvers if the computational budget is small. We demonstrate that this results in significant formal verification speed-ups.

Authors: De Palma, A; Behl, HS; Bunel, R; Torr, P; Kumar, MP

• Publication status: Published
• Peer review status: Peer reviewed
• Publisher: Open Review
• Journal: Proceedings of the ICLR 2021 Conference
• Pages: 126
• Publication date: 2021-01-14
• Acceptance date: 2021-01-12
• Event title: International Conference on Learning Representations (ICLR), 2021
• Event location: Online
• Event website: https://iclr.cc/Conferences/2021/Dates
• Event start date: 2021-05-04
• Event end date: 2021-05-08
• Language: English
• Keywords: FFR