arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2401.10310cs.LGcs.AIcs.CC
收藏

社会与司法约束下深度学习的数学算法设计:算法透明度要求

Mathematical Algorithm Design for Deep Learning under Societal and Judicial Constraints: The Algorithmic Transparency Requirement

Technical University of Munich(慕尼黑工业大学);Munich Center for Quantum Science and Technology (MCQST)(慕尼黑量子科学与技术中心);Ruhr-Universität Bochum(波鸿鲁尔大学);Ludwig-Maximilians-Universität München (LMU Munich)(慕尼黑大学);Munich Center for Machine Learning (MCML)(慕尼黑机器学习中心)

Holger Boche, Adalbert Fono, Gitta Kutyniok

AI总结:

本文提出数学框架分析算法透明实现的可行性,并比较 Turing 机与 Blum-Shub-Smale 机在逆问题深度学习中的可信性。

摘要:

深度学习在可信性方面仍存在缺陷,可信性描述的是一种可理解、公平、安全且可靠的方法。为降低人工智能的潜在风险,监管指南已提出与可信性相关的明确义务,例如《欧洲人工智能法案》。因此,一个核心问题是可信深度学习能够在多大程度上实现。确立构成可信性的上述属性,要求能够追溯影响算法计算的因素,即算法实现必须是透明的。受当前深度学习模型的发展使计算技术必须发生变革这一观察所推动,我们推导出一个数学框架,使我们能够分析在某种计算模型中实现透明实现是否可行。我们示例性地应用该可信性框架,分析分别由 Turing 机和 Blum-Shub-Smale 机表示的数字与模拟计算模型中用于逆问题的深度学习方法。基于已有结果,我们发现 Blum-Shub-Smale 机在相当一般的条件下有潜力为逆问题建立可信求解器,而 Turing 机无法在同等程度上保证可信性。

英文摘要:

Deep learning still has drawbacks in terms of trustworthiness, which describes a comprehensible, fair, safe, and reliable method. To mitigate the potential risk of AI, clear obligations associated to trustworthiness have been proposed via regulatory guidelines, e.g., in the European AI Act. Therefore, a central question is to what extent trustworthy deep learning can be realized. Establishing the described properties constituting trustworthiness requires that the factors influencing an algorithmic computation can be retraced, i.e., the algorithmic implementation is transparent. Motivated by the observation that the current evolution of deep learning models necessitates a change in computing technology, we derive a mathematical framework which enables us to analyze whether a transparent implementation in a computing model is feasible. We exemplarily apply our trustworthiness framework to analyze deep learning approaches for inverse problems in digital and analog computing models represented by Turing and Blum-Shub-Smale Machines, respectively. Based on previous results, we find that Blum-Shub-Smale Machines have the potential to establish trustworthy solvers for inverse problems under fairly general conditions, whereas Turing machines cannot guarantee trustworthiness to the same degree.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑