Modeling the Second Player in Distributionally Robust Optimization

by   Paul Michel, et al.

Distributionally robust optimization (DRO) provides a framework for training machine learning models that are able to perform well on a collection of related data distributions (the "uncertainty set"). This is done by solving a min-max game: the model is trained to minimize its maximum expected loss among all distributions in the uncertainty set. While careful design of the uncertainty set is critical to the success of the DRO procedure, previous work has been limited to relatively simple alternatives that keep the min-max optimization problem exactly tractable, such as f-divergence balls. In this paper, we argue instead for the use of neural generative models to characterize the worst-case distribution, allowing for more flexible and problem-specific selection of the uncertainty set. However, while simple conceptually, this approach poses a number of implementation and optimization challenges. To circumvent these issues, we propose a relaxation of the KL-constrained inner maximization objective that makes the DRO problem more amenable to gradient-based optimization of large scale generative models, and develop model selection heuristics to guide hyper-parameter search. On both toy settings and realistic NLP tasks, we find that the proposed approach yields models that are more robust than comparable baselines.


page 1

page 2

page 3

page 4


Specifying and Solving Robust Empirical Risk Minimization Problems Using CVXPY

We consider robust empirical risk minimization (ERM), where model parame...

New complexity results and algorithms for min-max-min robust combinatorial optimization

In this work we investigate the min-max-min robust optimization problem ...

Bayesian Optimization for Min Max Optimization

A solution that is only reliable under favourable conditions is hardly a...

Max-Min Off-Policy Actor-Critic Method Focusing on Worst-Case Robustness to Model Misspecification

In the field of reinforcement learning, because of the high cost and ris...

Robust Reinforcement Learning through Efficient Adversarial Herding

Although reinforcement learning (RL) is considered the gold standard for...

Large-Scale Methods for Distributionally Robust Optimization

We propose and analyze algorithms for distributionally robust optimizati...

Please sign up or login with your details

Forgot password? Click here to reset