Adaptation of XAI to Auto-tuning for Numerical Libraries

TL;DR

This paper explores how Explainable AI can be adapted to improve the transparency and effectiveness of auto-tuning processes in numerical libraries, addressing challenges in AI model explainability within performance optimization.

Contribution

It introduces methods for integrating XAI into AI-driven auto-tuning for numerical computations, enhancing interpretability in performance parameter tuning and sparse algorithms.

Findings

XAI techniques improve transparency in auto-tuning processes

Enhanced interpretability aids in performance optimization

Application to accuracy-guaranteed and sparse algorithms

Abstract

Concerns have arisen regarding the unregulated utilization of artificial intelligence (AI) outputs, potentially leading to various societal issues. While humans routinely validate information, manually inspecting the vast volumes of AI-generated results is impractical. Therefore, automation and visualization are imperative. In this context, Explainable AI (XAI) technology is gaining prominence, aiming to streamline AI model development and alleviate the burden of explaining AI outputs to users. Simultaneously, software auto-tuning (AT) technology has emerged, aiming to reduce the man-hours required for performance tuning in numerical calculations. AT is a potent tool for cost reduction during parameter optimization and high-performance programming for numerical computing. The synergy between AT mechanisms and AI technology is noteworthy, with AI finding extensive applications in AT. However, applying AI to AT mechanisms introduces challenges in AI model explainability. This research focuses on XAI for AI models when integrated into two different processes for practical numerical computations: performance parameter tuning of accuracy-guaranteed numerical calculations and sparse iterative algorithm.