Robust Deep Learning as Optimal Control: Insights and Convergence Guarantees
The fragility of deep neural networks to adversarially-chosen inputs has motivated the need to revisit deep learning algorithms. Including adversarial examples during training is a popular defense mechanism against adversarial attacks. This mechanism can be formulated as a min-max optimization problem, where the adversary seeks to maximize the loss function using an iterative first-order algorithm while the learner attempts to minimize it. However, finding adversarial examples in this way causes excessive computational overhead during training. By interpreting the min-max problem as an optimal control problem, it has recently been shown that one can exploit the compositional structure of neural networks in the optimization problem to improve the training time significantly. In this paper, we provide the first convergence analysis of this adversarial training algorithm by combining techniques from robust optimal control and inexact oracle methods in optimization. Our analysis sheds light on how the hyperparameters of the algorithm affect the its stability and convergence. We support our insights with experiments on a robust classification problem.
Code (0)
등록된 구현이 없습니다.
Tasks
Deep LearningRobust classificationSimilar Papers 제목 키워드 기반
Reinforcement Learning for Reachability: Guaranteeing Asymptotic Optimality
Reinforcement learning (RL) for reachability specifications is fundamental in sequential decision-making, yet theoretical guarantees remain less explored. A recent work achieves asymptotic convergence to optimal policies…
Reinforcement LearningConvergence Guarantees of Policy Optimization Methods for Markovian Jump Linear Systems
Recently, policy optimization for control purposes has received renewed attention due to the increasing interest in reinforcement learning. In this paper, we investigate the convergence of policy optimization for quadrat…
Reinforcement LearningOn the Convergence of Jacobian-Free Backpropagation for Optimal Control Problems with Implicit Hamiltonians
Optimal feedback control with implicit Hamiltonians poses a fundamental challenge for learning-based value function methods due to the absence of closed-form optimal control laws. Recent work~\cite{gelphman2025end} intro…
Optimally Managing the Impacts of Convergence Tolerance for Distributed Optimal Power Flow
The future power grid may rely on distributed optimization to determine the set-points for huge numbers of distributed energy resources. There has been significant work on applying distributed algorithms to optimal power…
Distributed OptimizationAccelerated Learning with Robustness to Adversarial Regressors
High order momentum-based parameter update algorithms have seen widespread applications in training machine learning models. Recently, connections with variational approaches have led to the derivation of new learning al…