Using Parameter Sensitivity and Interdependence to Predict Model Scope and Falsifiability
Using Parameter Sensitivity and Interdependence to Predict Model Scope and Falsifiability
复制标题
使用参数敏感性和相互依赖性来预测模型范围和可证伪性
DOI：
--
复制
发表时间：
1996
期刊：
影响因子：
--
通讯作者：
V. DeBrunner
中科院分区：
文献类型：
--
作者：
Shu Chen Li;S. Lewandowsky;V. DeBrunner
模型实用性的一个重要标准是它的范围，即预测各种结果的能力。如果没有广泛的数据拟合，范围往往难以确定。例如，J. E. Cutting，N. Bruno，N. P.布雷迪和C.摩尔（1992）通过拟合许多从底层函数任意生成的数据集，比较了感知视觉深度的两个模型。然后，他们将范围定义为模型可以考虑的功能数量。我们提出了一种替代技术的范围评估，是基于模型的参数的行为分析，并不需要大量的数据拟合。该技术检查模型参数之间的整体相互依赖性和它们的敏感性之间的比率，我们发现这与模型的范围成反比。
One important criterion for a model's utility is its scope, the ability to predict a wide range of results. Scope is often difficult to ascertain without extensive data fitting. For example, J. E. Cutting, N. Bruno, N. P. Brady, and C. Moore (1992) compared 2 models of perceived visual depth by fitting many data sets that were arbitrarily generated from underlying functions. They then defined scope as the number of functions a model could account for. We present an alternative technique for scope evaluation that is based on analysis of the behavior of a model's parameters and does not require extensive data fitting. The technique examines the ratio between the overall interdependence among model parameters and their sensitivity, which we show to be inversely related to a model's scope.
感知的范式和模糊逻辑模型仍然存在并且很好。
DOI：
10.1037//0096-3445.122.1.115
发表时间：
1993
期刊：
Journal of experimental psychology. General
影响因子：
--
作者：
Massaro,DW;Cohen,MM
通讯作者：
Cohen,MM