Pith. sign in

REVIEW 9 cited by

Neural Network Verification with Branch-and-Bound for General Nonlinearities

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2405.21063 v3 pith:G6AKPF5T submitted 2024-05-31 cs.LG cs.AI

classification cs.LGcs.AI
keywords verificationgeneralbranchinggenbablinearneuralnonlinearbranch-and-bound
verification ladder T0 review T1 audit T2 compute T3 formal

Signed reviews

No signed human review yet.

0 comments
abstract

Branch-and-bound (BaB) is among the most effective techniques for neural network (NN) verification. However, existing works on BaB for NN verification have mostly focused on NNs with piecewise linear activations, especially ReLU networks. In this paper, we develop a general framework, named GenBaB, to conduct BaB on general nonlinearities to verify NNs with general architectures, based on linear bound propagation for NN verification. To decide which neuron to branch, we design a new branching heuristic which leverages linear bounds as shortcuts to efficiently estimate the potential improvement after branching. To decide nontrivial branching points for general nonlinear functions, we propose to pre-optimize branching points, which can be efficiently leveraged during verification with a lookup table. We demonstrate the effectiveness of our GenBaB on verifying a wide range of NNs, including NNs with activation functions such as Sigmoid, Tanh, Sine and GeLU, as well as NNs involving multi-dimensional nonlinear operations such as multiplications in LSTMs and Vision Transformers. Our framework also allows the verification of general nonlinear computation graphs and enables verification applications beyond simple NNs, particularly for AC Optimal Power Flow (ACOPF). GenBaB is part of the latest $\alpha$,$\beta$-CROWN, the winner of the 4th and the 5th International Verification of Neural Networks Competition (VNN-COMP 2023 and 2024). Code for reproducing the experiments is available at https://github.com/shizhouxing/GenBaB.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 9 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Lipschitz-Based Robustness Certification Under Floating-Point Execution

    cs.LG 2026-03 conditional novelty 7.0 of 10 partial

    Lipschitz-based robustness certificates that assume real arithmetic can be unsound under floating-point execution; a formal FP-aware theory and certifier close that gap for dense ReLU networks.

  2. Of Good Demons and Bad Angels: Guaranteeing Safe Control under Finite Precision

    eess.SY 2025-07 conditional novelty 7.0 of 10

    A dL/dGL-based method that verifies infinite-horizon safety of neural network controllers under bounded finite-precision perturbations and synthesizes sound mixed-precision fixed-point implementations.

  3. Adaptive Branch-and-Bound Tree Exploration for Neural Network Verification

    cs.LG 2025-05 conditional novelty 6.0 of 10

    ABONN orders branch-and-bound sub-problems by a counterexample potentiality score and reports speedups of up to 15.2x on MNIST and 24.7x on CIFAR-10 over a naive branch-and-bound baseline.

  4. Neural Network Certification Informed Power System Transient Stability Preventive Control with Renewable Energy

    eess.SY 2024-11 conditional novelty 6.0 of 10

    The paper integrates a deep belief network surrogate into optimal power flow and uses α,β-CROWN certification to iteratively adjust the transient stability safety margin under uncertainty.

  5. A General Framework for Property-Driven Machine Learning

    cs.LG 2025-05 conditional novelty 5.0 of 10

    A unified training objective generalizing adversarial training and differentiable-logic constraints, demonstrated on image classification and a drone controller.

  6. Learning Ensembles of Vision-based Safety Control Filters

    cs.LG 2024-12 conditional novelty 5.0 of 10

    Ensembles of vision-based safety filters with diverse backbones and aggregation methods improve safe/unsafe classification accuracy over individual models on the DeepAccident dataset.

  7. Efficient Neural Network Verification via Order Leading Exploration of Branch-and-Bound Trees

    cs.LG 2025-07 conditional novelty 4.0 of 10

    Reordering branch-and-bound sub-problems by a counterexample-potentiality heuristic accelerates neural network verification, especially for falsified instances.

  8. The Fifth International Verification of Neural Networks Competition (VNN-COMP 2024): Summary and Results

    cs.LG 2024-12 conditional novelty 4.0 of 10

    The 2024 VNN-COMP report documents that GPU-accelerated linear bound propagation with branch-and-bound, led by α,β-CROWN, dominated both the regular and extended tracks.

  9. Creating a Formally Verified Neural Network for Autonomous Navigation: An Experience Report

    cs.LO 2024-11 conditional novelty 4.0 of 10

    A case study shows that differentiable-logic training improves local robustness of a small path-centring network, but current verifiers fail on the regression architecture and the title's 'formally verified' claim is ...

Pith tools