Are Transformers More Robust? Towards Exact Robustness Verification for Transformers
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2022
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866915040820461568 |
|---|---|
| author | Liao, Brian Hsuan-Cheng Cheng, Chih-Hong Esen, Hasan Knoll, Alois |
| author_facet | Liao, Brian Hsuan-Cheng Cheng, Chih-Hong Esen, Hasan Knoll, Alois |
| contents | As an emerging type of Neural Networks (NNs), Transformers are used in many domains ranging from Natural Language Processing to Autonomous Driving. In this paper, we study the robustness problem of Transformers, a key characteristic as low robustness may cause safety concerns. Specifically, we focus on Sparsemax-based Transformers and reduce the finding of their maximum robustness to a Mixed Integer Quadratically Constrained Programming (MIQCP) problem. We also design two pre-processing heuristics that can be embedded in the MIQCP encoding and substantially accelerate its solving. We then conduct experiments using the application of Land Departure Warning to compare the robustness of Sparsemax-based Transformers against that of the more conventional Multi-Layer-Perceptron (MLP) NNs. To our surprise, Transformers are not necessarily more robust, leading to profound considerations in selecting appropriate NN architectures for safety-critical domain applications. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2202_03932 |
| institution | arXiv |
| publishDate | 2022 |
| record_format | arxiv |
| spellingShingle | Are Transformers More Robust? Towards Exact Robustness Verification for Transformers Liao, Brian Hsuan-Cheng Cheng, Chih-Hong Esen, Hasan Knoll, Alois Machine Learning Artificial Intelligence As an emerging type of Neural Networks (NNs), Transformers are used in many domains ranging from Natural Language Processing to Autonomous Driving. In this paper, we study the robustness problem of Transformers, a key characteristic as low robustness may cause safety concerns. Specifically, we focus on Sparsemax-based Transformers and reduce the finding of their maximum robustness to a Mixed Integer Quadratically Constrained Programming (MIQCP) problem. We also design two pre-processing heuristics that can be embedded in the MIQCP encoding and substantially accelerate its solving. We then conduct experiments using the application of Land Departure Warning to compare the robustness of Sparsemax-based Transformers against that of the more conventional Multi-Layer-Perceptron (MLP) NNs. To our surprise, Transformers are not necessarily more robust, leading to profound considerations in selecting appropriate NN architectures for safety-critical domain applications. |
| title | Are Transformers More Robust? Towards Exact Robustness Verification for Transformers |
| topic | Machine Learning Artificial Intelligence |
| url | https://arxiv.org/abs/2202.03932 |