Abstract:In this paper, stability and sensitivity properties of a class of parametric constrained optimization problem, whose feasible region is defined by a set-valued inclusion, are investigated through the associated optimal value function. Set-valued inclusions are a kind of constraint system, which naturally emerges in contexts requiring the robust fulfilment of traditional cone constraints, where data are affected by uncertain elements having a non stochastic nature, or in (MPEC) as a vector equilibrium constraint, where feasible solutions are intended as equilibrium point in a strong sense. Under proper convexity assumptions on the objective function and the constraining set-valued term, combined with a global qualification condition, a class of parametric optimization problems is singled out, which displays a global Lipschitz behaviour. By employing recent results of variational analysis, elements for a sensitivity analysis of this class of problems are provided via exact subgradient formulae for the optimal value function. Further consequences of the stability behaviour are explored in terms of problem calmness and viability of penalization techniques.
What problem does this paper attempt to address?
This paper attempts to solve the stability and sensitivity analysis of parametric optimization problems with set - valued inclusion constraints. Specifically, the author studies a special class of optimization problems whose feasible regions are defined by set - valued inclusions. Such problems have important applications in areas such as robust optimization and extensions of mathematical programming.
### Main problems and objectives
1. **Stability analysis**: Study the behavior of the optimal value function (marginal function) when parameters change. In particular, the author focuses on the impact of such changes on the solutions of the problem and how to quantify these impacts.
2. **Sensitivity analysis**: Provide the sub - gradient formula of the optimal value function to better understand the impact of parameter changes on the problem. This helps to evaluate the behavior of the problem under different parameters and provides theoretical support for decision - making in practical applications.
### Research methods and technical means
- **Set - valued inclusion constraints**: The paper introduces the concept of set - valued inclusion, which is a new form of constraint and can describe complex constraint conditions more flexibly.
- **Convexity assumptions**: Through appropriate convexity assumptions and global qualification conditions, the author proves that this class of problems has the global Lipschitz property.
- **Tools of variational analysis**: Utilize the latest results in variational analysis to provide an accurate sub - gradient formula for sensitivity analysis.
- **Continuous selection theorem**: Through the continuous selection theorem, ensure the existence and continuous dependence of solutions.
### Main conclusions
- Under appropriate convexity and qualification conditions, the optimal value function is locally Lipschitz continuous, which means it has good stability when parameters change.
- Provide an accurate sub - gradient formula for sensitivity analysis, which helps to understand and predict the impact of parameter changes on the problem solutions.
- Explore the "calmness" of the problem and the effectiveness of penalty techniques, and these properties are of great significance for algorithm design and optimization in practical applications.
### Formula presentation
- **Feasible region mapping**:
\[
R(p)=F^{-1}(p, \cdot)(C)=\{x \in X \mid F(p, x) \subseteq C\}
\]
- **Optimal value function**:
\[
\text{val}(p)=\inf_{x \in R(p)} \phi(p, x)
\]
- **Sub - gradient formula**:
\[
\partial \text{val}(\bar{p})=\left\{p^{*}+q^{*} \mid\left(p^{*}, x^{*}\right) \in \partial \phi(\bar{p}, \bar{x}),\left(q^{*}, -x^{*}\right) \in \text{cone } \partial \text{exc}(F(\cdot); C)(\bar{p}, \bar{x})\right\}
\]
Through these studies, the paper provides in - depth theoretical analysis and practical computational tools for optimization problems with set - valued inclusion constraints.