arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2106.03428cs.LG
收藏

通过损失函数与调节器的比较实现可解释机器学习的自动化

Automation for Interpretable Machine Learning Through a Comparison of Loss Functions to Regularisers

University of Southampton(南安普顿大学);The Alan Turing Institute(艾伦·图灵研究所);Silverstream Technologies Ltd(Silverstream科技有限公司)

A. I. Parkes, J. Camilleri, D. A. Hudson, A. J. Sobey

摘要:

为了增加机器学习的普及,它需要自动化。自动化具有成本效益,因为它允许专家花费更少的时间来调整方法,从而缩短开发时间。然而,尽管这种自动化产生了高度准确的架构,但它们可能是不可解释的,充当“黑盒子”,产生较低的常规错误,但无法对底层的输入-输出关系(地面事实)进行建模。本文探讨了在机器学习回归自动化中使用拟合中位误差度量,使用演化计算以提高对地面真实值的近似。当与传统的误差测量一起使用时,它通过将学习到的输入输出关系正则化到条件中值来提高可解释性。它与传统的正则化器进行了比较,以说明使用拟合中位误差产生回归神经网络,该回归神经网络对更一致的输入-输出关系进行建模。所考虑的问题是船舶功率预测使用的节油空气润滑系统,这是高度随机的性质。针对拟合中值误差进行优化的网络显示出更一致地近似地面实况,而不会牺牲传统的Minkowski-r误差值。

英文摘要:

To increase the ubiquity of machine learning it needs to be automated. Automation is cost-effective as it allows experts to spend less time tuning the approach, which leads to shorter development times. However, while this automation produces highly accurate architectures, they can be uninterpretable, acting as `black-boxes' which produce low conventional errors but fail to model the underlying input-output relationships -- the ground truth. This paper explores the use of the Fit to Median Error measure in machine learning regression automation, using evolutionary computation in order to improve the approximation of the ground truth. When used alongside conventional error measures it improves interpretability by regularising learnt input-output relationships to the conditional median. It is compared to traditional regularisers to illustrate that the use of the Fit to Median Error produces regression neural networks which model more consistent input-output relationships. The problem considered is ship power prediction using a fuel-saving air lubrication system, which is highly stochastic in nature. The networks optimised for their Fit to Median Error are shown to approximate the ground truth more consistently, without sacrificing conventional Minkowski-r error values.

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。

↑