We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:

Download:

Current browse context:

cs.LG

Change to browse by:

References & Citations

DBLP - CS Bibliography

Bookmark

(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Computer Science > Machine Learning

Title: On Certifying Robust Models by Polyhedral Envelope

Abstract: Certifying neural networks enables one to offer guarantees on a model's robustness. In this work, we use linear approximation to obtain an upper and lower bound of the model's output when the input data is perturbed within a predefined adversarial budget. This allows us to bound the adversary-free region in the data neighborhood by a polyhedral envelope, and calculate robustness guarantees based on this geometric approximation. Compared with existing methods, our approach gives a finer-grain quantitative evaluation of a model's robustness. Therefore, the certification method can not only obtain better certified bounds than the state-of-the-art techniques given the same adversarial budget but also derives a faster search scheme for the optimal adversarial budget. Furthermore, we introduce a simple regularization scheme based on our method that enables us to effectively train robust models.
Subjects: Machine Learning (cs.LG); Machine Learning (stat.ML)
Cite as: arXiv:1912.04792 [cs.LG]
  (or arXiv:1912.04792v1 [cs.LG] for this version)

Submission history

From: Chen Liu [view email]
[v1] Tue, 10 Dec 2019 16:05:20 GMT (5956kb,D)

Link back to: arXiv, form interface, contact.