BriefGPT.xyz
Apr, 2024
SHIELD:可解释人工智能的正则化技术
SHIELD: A regularization technique for eXplainable Artificial Intelligence
HTML
PDF
Iván Sevillano-García, Julián Luengo, Francisco Herrera
TL;DR
介绍了一种名为SHIELD(选择性隐藏输入评估学习动力学)的可解释人工智能正则化技术,通过隐藏部分输入数据并评估预测结果的差异,改善模型质量、提高解释性,并在基准数据集上验证了其有效性,为开发透明可靠的人工智能正则化技术开辟了有前途的道路。
Abstract
As
artificial intelligence
systems become integral across domains, the demand for
explainability
grows. While the effort by the scientific community is focused on obtaining a better explanation for the model, it
→