Model Selection, Simplicity, and Scientific Inference

Model Selection, Simplicity, and Scientific Inference

The Akaike Information Criterion can be a valuable tool of scientific inference. This statistic, or any other statistical method for that matter, cannot, however, be the whole of scientific methodology. In this paper some of the limitations of Akaikean statistical methods are discussed. It is argued that the full import of empirical evidence is realized only by adopting a richer ideal of empiri...

Inference of Regular Languages using Model Simplicity

We describe an approach that is related to a number of existing algorithms for the inference of a regular language from a set of positive (and optionally also negative) examples. Variations on this approach provide a family of algorithms that attempt to minimise the complexity of a description of the example data in terms of a finite state automaton model. Experiments using a standard set of sm...

Simplicity and Unification in Model Selection

This chapter examines four solutions to the problem of many models, and finds some fault or limitation with all of them except the last. The first is the naïve empiricist view that best model is the one that best fits the data. The second is based on Popper’s falsificationism. The third approach is to compare models on the basis of some kind of trade off between fit and simplicity. The fourth i...

Inference Principles and Model Selection

The core problem of statistics and machine learning addresses the question how can we efficiently find a statistical model to describe empirical data. Classical statistical approaches to solve this problem have been complemented during the last 15 years by Neural Computation, a very promising strategy to data analysis. The Dagstuhl seminar on “Inference Principles and Model Selection” — the fou...

Model Selection and Multimodel Inference