Abstract:The recent results of An, Luan, and Yen [Differential stability in convex optimization via generalized polyhedrality. Vietnam J. Math. <a class="link-external link-https" href="https://-doi.org/10.1007/s10013-024-00721-y" rel="external noopener nofollow">this https URL</a>] on differential stability of parametric optimization problems described by proper generalized polyhedral convex functions and generalized polyhedral convex set-valued maps are analyzed, developed, and sharpened in this paper. Namely, keeping the Hausdorff locally convex topological vector spaces setting, we clarify the relationships between the upper estimates and lower estimates for the subdifferential and the singular subdifferential of the optimal value function. As shown by an example, the lower estimates can be strict. But, surprisingly, each upper estimate is an equality. Thus, exact formulas for the subdifferential and the singular subdifferential under consideration are obtained. In addition, it is proved that each subdifferential upper estimate coincides with the corresponding lower estimate if either the objective function or the constraint set-valued map is polyhedral convex.
What problem does this paper attempt to address?
The problem that this paper attempts to solve is the differential stability analysis of a class of convex optimization problems. Specifically, the authors aim to analyze, develop and deepen the results obtained by An, Luan and Yen in previous studies. These results are related to the differential stability of parameter - optimization problems described by generalized polyhedral convex functions and generalized polyhedral convex set - valued mappings.
### Main problems
1. **Relationship of differential stability**: The authors hope to clarify the relationship between the upper - bound estimates and the lower - bound estimates of the subdifferential and the singular subdifferential of the optimal value function.
2. **Strictness verification**: Verify through examples that the lower - bound estimate can be strict, and each upper - bound estimate is actually an equality.
3. **Exact formulas**: Obtain the exact formulas of the subdifferential and the singular subdifferential.
4. **Equality in special cases**: Prove that when the objective function or the constraint set - valued mapping is polyhedral convex, each subdifferential upper - bound estimate is equal to the corresponding lower - bound estimate.
### Specific problems
- **Question 1**: Are there some examples showing that each estimate provided by Theorems 2.4 and 2.5 can be strict?
- **Question 2**: Are the sets \( B \) and \( B^\infty \) always closed?
- **Question 3**: If \( \partial \mu(\bar{x})=A \), then is \( B = A \)?
- **Question 4**: If \( \partial^\infty \mu(\bar{x})=A^\infty \), then is \( B^\infty = A^\infty \)?
- **Question 5**: For any parameterized generalized polyhedral convex optimization problem of the form (Px), do equations (19) and (20) hold for any \( \bar{x} \in X \) satisfying \( \mu(\bar{x}) \in \mathbb{R} \) and \( \bar{y} \in M(\bar{x}) \)?
### Solutions
- **Construct counterexamples**: Prove that some estimates can be strict by constructing appropriate examples.
- **Theoretical derivation**: Use existing theorems and lemmas, such as Theorem 2.2 and Theorem 2.3, to derive new conclusions.
- **Analysis of special cases**: Analyze the situation when the objective function or the constraint set - valued mapping is polyhedral convex, and prove that the upper - and lower - bound estimates are equal at this time.
### Conclusion
Through a series of theoretical derivations and example verifications, the authors have solved the above problems and obtained new results regarding the subdifferential and the singular subdifferential. In particular, they have proved that in some special cases, the upper - and lower - bound estimates of the subdifferential are equal, thus providing a deeper understanding of the differential stability of convex optimization problems.