Robust priors for regularized regression
This provides a principled way to interpolate between models of differing complexity for data analysis and machine learning applications, though it appears incremental in extending existing penalized regression methods.
The paper tackled the problem of inappropriate zero-centered priors in penalized regression by constructing non-zero priors inspired by human decision heuristics, resulting in robust and interpretable solutions with excellent worst-case performance across decision, classification, and simulated brain imaging tasks.
Induction benefits from useful priors. Penalized regression approaches, like ridge regression, shrink weights toward zero but zero association is usually not a sensible prior. Inspired by simple and robust decision heuristics humans use, we constructed non-zero priors for penalized regression models that provide robust and interpretable solutions across several tasks. Our approach enables estimates from a constrained model to serve as a prior for a more general model, yielding a principled way to interpolate between models of differing complexity. We successfully applied this approach to a number of decision and classification problems, as well as analyzing simulated brain imaging data. Models with robust priors had excellent worst-case performance. Solutions followed from the form of the heuristic that was used to derive the prior. These new algorithms can serve applications in data analysis and machine learning, as well as help in understanding how people transition from novice to expert performance.