Abstract:We study a dynamic stochastic control problem subject to Knightian uncertainty with multi-objective (vector-valued) criteria. Assuming the preferences across expected multi-loss vectors are represented by a given, yet general, preorder, we address the model uncertainty by adopting a robust or minimax perspective, minimizing expected loss across the worst-case model. For loss functions taking real (or scalar) values, there is no ambiguity in interpreting supremum and infimum. In contrast to the scalar case, major challenges for multi-loss control problems include properly defining and interpreting the notions of supremum and infimum, and addressing the non-uniqueness of these suprema and infima. To deal with these, we employ the notion of an ideal point vector-valued supremum for the robust part of the problem, while we view the control part as a multi-objective (or vector) optimization problem. Using a set-valued framework, we derive both a weak and strong version of the dynamic programming principle (DPP) or Bellman equations by taking the value function as the collection of all worst expected losses across all feasible actions. The weak version of Bellman's principle is proved under minimal assumptions. To establish a stronger version of DPP, we introduce the rectangularity property with respect to a general preorder. We also further study a particular, but important, case of component-wise partial order of vectors, for which we additionally derive DPP under a different set-valued notion for the value function, the so-called upper image of the multi-objective problem. Finally, we provide illustrative examples motivated by financial problems. These results will serve as a foundation for addressing time-inconsistent problems subject to model uncertainty through the lens of a set-valued framework, as well as for studying multi-portfolio allocation problems under model uncertainty.

Treatment of Set-Valued Robustness via Separation and Scalarization

Scalarizations and Optimality of Constrained Set-Valued Optimization Using Improvement Sets and Image Space Analysis

Scalarisation-based risk concepts for robust multi-objective optimisation

Vector-valued robust stochastic control

Optimality conditions for robust weakly efficient solutions in uncertain optimization

Theoretical aspects of robust SVM optimization in Banach spaces and Nash equilibrium interpretation

Projection-based Robust Optimization with Symbolic Computation

Existence of Weak Efficient Solutions of Set-Valued Optimization Problems

Robust Mathematical Programming Problems Involving Vanishing Constraints via Strongly Invex Functions

The robust isolated calmness of spectral norm regularized convex matrix optimization problems

An unified scalarization and Ekeland's variational principle for partial set order

Decision space robustness for multi-objective integer linear programming

On the robust isolated calmness of a class of nonsmooth optimizations on Riemannian manifolds and its applications

Mixed-Integer Programming for a Class of Robust Submodular Maximization Problems

Smooth Robustness Measures for Symbolic Control Via Signal Temporal Logic

The Continuity and Convexity of a Nonlinear Scalarization Function with Applications in Set Optimization Problems Involving a Partial Order Relation

Roping in Uncertainty: Robustness and Regularization in Markov Games

The Price of Robustness

Affinely Adjustable Robust Linear Complementarity Problems