Research map / Learning theory and algorithms / Combinatorial and search optimization
Fast and Complete: Enabling Complete Neural Network Verification with Rapid and Massively Parallel Incomplete Verifiers
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| replace LP bounding with optimized LiRPA and batch-parallel splits | BaB neural network verification · CIFAR-10 Base Easy | time(s): 15.68 -> 11.86 | Helped |
| replace LP bounding with optimized LiRPA and batch-parallel splits | BaB neural network verification · CIFAR-10 Base Medium | time(s): 51.88 -> 42.04 | Helped |
| replace LP bounding with optimized LiRPA and batch-parallel splits | BaB neural network verification · CIFAR-10 Base Hard | time(s): 627.96 -> 633.85 | Hurt |
| replace LP bounding with optimized LiRPA and batch-parallel splits | BaB neural network verification · CIFAR-10 Wide | time(s): 510.55 -> 375.23 | Helped |
| replace LP bounding with optimized LiRPA and batch-parallel splits | BaB neural network verification · CIFAR-10 Deep | time(s): 230.06 -> 81.55 | Helped |
| add batch splits without optimizing LiRPA bounds (ablation) | BaBSR baseline · CIFAR-10 Base Easy | time(s): 522.48 -> 587.10 | Hurt |
| optimize LiRPA bounds (ablation) | LiRPA-based BaB verifier without batch splits · CIFAR-10 Base Easy | time(s): 522.48 -> 94.08 | Helped |
| add batch splits to optimized LiRPA (ablation) | Optimized LiRPA verifier · CIFAR-10 Base Easy | time(s): 94.08 -> 11.86 | Helped |
| add optimized LiRPA bounds and batch splits together (ablation) | BaBSR baseline · CIFAR-10 Base Hard | time(s): 2875.16 -> 633.85 | Helped |
| rely on LiRPA without LP feasibility checks | BaB verifier · complete NN verification | completeness | Hurt |
Closest papers by meaning
- Clip-and-Verify: Linear Constraint-Driven Domain Clipping for Accelerating Neural Network Verification (NeurIPS 2025 · similarity 0.89)
- SDP-CROWN: Efficient Bound Propagation for Neural Network Verification with Tightness of Semidefinite Programming (ICML 2025 · similarity 0.88)
- Scalable Neural Network Verification with Branch-and-bound Inferred Cutting Planes (NeurIPS 2024 · similarity 0.87)
- Dynamic Back-Substitution in Bound-Propagation-Based Neural Network Verification (AAAI 2025 · similarity 0.86)
- DeepBern-Nets: Taming the Complexity of Certifying Neural Networks Using Bernstein Polynomial Activations and Precise Bound Propagation (AAAI 2024 · similarity 0.85)
- Certified but Private: Scalable Zero-Knowledge Proofs for the Formal Verification of Neural Networks (NeurIPS 2026 · similarity 0.85)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.