Verifying Recurrent Neural Networks using Invariant Inference
This addresses the need for reliable verification of RNNs used in domains like natural language processing, but it appears incremental as it builds on existing verification techniques with a specific improvement.
The paper tackles the problem of verifying recurrent neural networks (RNNs) for reliability in critical systems by proposing an invariant inference approach, which reduces verification complexity and demonstrates orders-of-magnitude better performance than state-of-the-art methods in experiments.
Deep neural networks are revolutionizing the way complex systems are developed. However, these automatically-generated networks are opaque to humans, making it difficult to reason about them and guarantee their correctness. Here, we propose a novel approach for verifying properties of a widespread variant of neural networks, called recurrent neural networks. Recurrent neural networks play a key role in, e.g., natural language processing, and their verification is crucial for guaranteeing the reliability of many critical systems. Our approach is based on the inference of invariants, which allow us to reduce the complex problem of verifying recurrent networks into simpler, non-recurrent problems. Experiments with a proof-of-concept implementation of our approach demonstrate that it performs orders-of-magnitude better than the state of the art.