Faith: An Efficient Framework for Transformer Verification on GPUs
This work addresses performance bottlenecks in transformer verification for machine learning researchers and practitioners, offering incremental improvements in efficiency.
The paper tackles the problem of inefficient transformer verification on GPUs by proposing Faith, a framework that uses semantic-aware graph transformation, specialized kernel crafting, and expert-guided autotuning, resulting in an average speedup of 2.6x over state-of-the-art methods.
Transformer verification draws increasing attention in machine learning research and industry. It formally verifies the robustness of transformers against adversarial attacks such as exchanging words in a sentence with synonyms. However, the performance of transformer verification is still not satisfactory due to bound-centric computation which is significantly different from standard neural networks. In this paper, we propose Faith, an efficient framework for transformer verification on GPUs. We first propose a semantic-aware computation graph transformation to identify semantic information such as bound computation in transformer verification. We exploit such semantic information to enable efficient kernel fusion at the computation graph level. Second, we propose a verification-specialized kernel crafter to efficiently map transformer verification to modern GPUs. This crafter exploits a set of GPU hardware supports to accelerate verification specialized operations which are usually memory-intensive. Third, we propose an expert-guided autotuning to incorporate expert knowledge on GPU backends to facilitate large search space exploration. Extensive evaluations show that Faith achieves $2.1\times$ to $3.4\times$ ($2.6\times$ on average) speedup over state-of-the-art frameworks.