Quality experts often need to identify in software systems design defects, which are recurring design problems, that hinder development and maintenance. Consequently, several defect detection approaches and tools have...
详细信息
ISBN:
(纸本)9783540787426
Quality experts often need to identify in software systems design defects, which are recurring design problems, that hinder development and maintenance. Consequently, several defect detection approaches and tools have been proposed in the literature. However, we are not aware of any approach that defines and reifies the process of generating detection algorithms from the existing textual descriptions of defects. In this paper, we introduce an approach to automate the generation of detection algorithms from specifications written using a domain-specific language. The domain-specific is defined from a thorough domain analysis. We specify several design defects, generate automatically detection algorithms using templates, and validate the generated detection algorithms in terms of precision and recall on XERCES v2.7.0, an open-source object-oriented system.
This paper treats the task of designing optimization algorithms as an optimal control problem. Using regret as a metric for an algorithm's performance, we study the existence, uniqueness and consistency of regret-...
详细信息
This paper treats the task of designing optimization algorithms as an optimal control problem. Using regret as a metric for an algorithm's performance, we study the existence, uniqueness and consistency of regret-optimal algorithms. By providing first-order optimality conditions for the control problem, we show that regret-optimal algorithms must satisfy a specific structure in their dynamics which we show is equivalent to performing dual-preconditioned gradient descent on the value function generated by its regret. Using these optimal dynamics, we provide bounds on their rates of convergence to solutions of convex optimization problems. Though closed-form optimal dynamics cannot be obtained in general, we present fast numerical methods for approximating them, generating optimization algorithms which directly optimize their long-term regret. These are benchmarked against commonly used optimization algorithms to demonstrate their effectiveness.
暂无评论