LGAILODec 20, 2023

Towards Efficient Verification of Quantized Neural Networks

arXiv:2312.12679v223 citationsh-index: 13AAAI
Originality Incremental advance
AI Analysis

This work addresses the need for reliable verification in quantized networks, which are crucial for efficient on-device AI, but it is incremental as it builds on existing verification techniques.

The authors tackled the problem of formally verifying properties of quantized neural networks, achieving better scalability and efficiency than previous state-of-the-art methods.

Quantization replaces floating point arithmetic with integer arithmetic in deep neural network models, providing more efficient on-device inference with less power and memory. In this work, we propose a framework for formally verifying properties of quantized neural networks. Our baseline technique is based on integer linear programming which guarantees both soundness and completeness. We then show how efficiency can be improved by utilizing gradient-based heuristic search methods and also bound-propagation techniques. We evaluate our approach on perception networks quantized with PyTorch. Our results show that we can verify quantized networks with better scalability and efficiency than the previous state of the art.

Code Implementations1 repo
Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes