A Method for Controlling Extrapolation when Visualizing and Optimizing the Prediction Profiles of Statistical and Machine Learning Models
This work addresses a practical issue for users of JMP software in data analysis and modeling, offering an incremental improvement to existing visualization tools.
The authors tackled the problem of extrapolation in prediction profilers for statistical and machine learning models, presenting a method that helps users avoid invalid predictions and optimizes factor settings to prevent extrapolation, demonstrating in simulations and real-world examples that unconstrained optimizations often lead to extrapolated solutions.
We present a novel method for controlling extrapolation in the prediction profiler in the JMP software. The prediction profiler is a graphical tool for exploring high dimensional prediction surfaces for statistical and machine learning models. The profiler contains interactive cross-sectional views, or profile traces, of the prediction surface of a model. Our method helps users avoid exploring predictions that should be considered extrapolation. It also performs optimization over a constrained factor region that avoids extrapolation using a genetic algorithm. In simulations and real world examples, we demonstrate how optimal factor settings without constraint in the profiler are frequently extrapolated, and how extrapolation control helps avoid these solutions with invalid factor settings that may not be useful to the user.