Research map / Architectures and efficiency / Quantization
Scalable Verification of Quantized Neural Networks
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 |
|---|---|---|---|
| Apply dead-branch removal, minimum bitwidth allocation, and redundancy elimination | Boolector quantized-network verification · MNIST and Fashion-MNIST adversarial robustness | median runtime per sample: 8803 | 8789; 2798 | 3931; 5 | 90; 6927 | 6927; 3105 | 3474; 4 | 49 | Helped |
| Apply the proposed encoding improvements | Boolector quantized-network verification · MNIST and Fashion-MNIST adversarial robustness | solved instances | Helped |
| Add redundancy elimination (ablation) | Proposed QNN verification method · MNIST | solved instances: 316 (80.8%) -> 318 (81.3%) | Helped |
| Add minimum bitwidth allocation (ablation) | Proposed QNN verification method · MNIST | solved instances: 315 (80.6%) -> 318 (81.3%) | Helped |
| Add ReLU simplification (ablation) | Proposed QNN verification method · MNIST | solved instances: 88 (22.5%) -> 318 (81.3%) | Helped |
| Add abstract interpretation (ablation) | Proposed QNN verification method · MNIST | solved instances: 107 (27.4%) -> 318 (81.3%) | Helped |
| Add redundancy elimination (ablation) | Proposed QNN verification method · MNIST | cumulative runtime: 7.7 h -> 7.9 h | Mixed |
| Add minimum bitwidth allocation (ablation) | Proposed QNN verification method · MNIST | cumulative runtime: 5.1 h -> 7.9 h | Mixed |
Closest papers by meaning
- Towards Efficient Verification of Quantized Neural Networks (AAAI 2024 · similarity 0.92)
- Training Quantized Nets: A Deeper Understanding (NeurIPS 2017 · similarity 0.88)
- Certified but Private: Scalable Zero-Knowledge Proofs for the Formal Verification of Neural Networks (NeurIPS 2026 · similarity 0.87)
- Verifying Properties of Binary Neural Networks Using Sparse Polynomial Optimization (ICLR 2025 · similarity 0.86)
- VNN: Verification-Friendly Neural Networks with Hard Robustness Guarantees (ICML 2024 · similarity 0.86)
- Training Verification-Friendly Neural Networks via Neuron Behavior Consistency (AAAI 2025 · 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.