A fast saddle-point dynamical system approach to robust deep learning

Yasaman Esfandiari, Aditya Balu, Keivan Ebrahimi, Umesh Vaidya, Nicola Elia, Soumik Sarkar

Research output: Contribution to journalArticlepeer-review

3 Scopus citations


Recent focus on robustness to adversarial attacks for deep neural networks produced a large variety of algorithms for training robust models. Most of the effective algorithms involve solving the min–max optimization problem for training robust models (min step) under worst-case attacks (max step). However, they often suffer from high computational cost from running several inner maximization iterations (to find an optimal attack) inside every outer minimization iteration. Therefore, it becomes difficult to readily apply such algorithms for moderate to large size real world data sets. To alleviate this, we explore the effectiveness of iterative descent–ascent algorithms where the maximization and minimization steps are executed in an alternate fashion to simultaneously obtain the worst-case attack and the corresponding robust model. Specifically, we propose a novel discrete-time dynamical system-based algorithm that aims to find the saddle point of a min–max optimization problem in the presence of uncertainties. Under the assumptions that the cost function is convex and uncertainties enter concavely in the robust learning problem, we analytically show that our algorithm converges asymptotically to the robust optimal solution under a general adversarial budget constraints as induced by ℓp norm, for 1≤p≤∞. Based on our proposed analysis, we devise a fast robust training algorithm for deep neural networks. Although such training involves highly non-convex robust optimization problems, empirical results show that the algorithm can achieve significant robustness compared to other state-of-the-art robust models on benchmark data sets.

Original languageEnglish (US)
Pages (from-to)33-44
Number of pages12
JournalNeural Networks
StatePublished - Jul 2021

Bibliographical note

Funding Information:
This work was partially supported by NSF, USA CAREER Grant (CNS#1845969).

Publisher Copyright:
© 2021 Elsevier Ltd


  • Adversarial training
  • Robust deep learning
  • Robust optimization
  • Neural Networks, Computer
  • Pattern Recognition, Automated/methods
  • Algorithms
  • Deep Learning
  • Databases, Factual

PubMed: MeSH publication types

  • Journal Article


Dive into the research topics of 'A fast saddle-point dynamical system approach to robust deep learning'. Together they form a unique fingerprint.

Cite this