Tradeoffs in Constructing and Evaluating Temporal Influence Diagrams

TL;DR

This paper explores the tradeoffs in constructing and evaluating Temporal Influence Diagrams to balance accuracy, complexity, and computational feasibility in modeling dynamic systems.

Contribution

It introduces three approaches for TID construction that balance data, complexity, and predictive accuracy, along with evaluation methods for these tradeoffs.

Findings

Tailoring networks to data improves relevance.

Modeling key variables reduces complexity.

Model selection balances accuracy and overfitting.

Abstract

This paper addresses the tradeoffs which need to be considered in reasoning using probabilistic network representations, such as Influence Diagrams (IDs). In particular, we examine the tradeoffs entailed in using Temporal Influence Diagrams (TIDs) which adequately capture the temporal evolution of a dynamic system without prohibitive data and computational requirements. Three approaches for TID construction which make different tradeoffs are examined: (1) tailoring the network at each time interval to the data available (rather then just copying the original Bayes Network for all time intervals); (2) modeling the evolution of a parsimonious subset of variables (rather than all variables); and (3) model selection approaches, which seek to minimize some measure of the predictive accuracy of the model without introducing too many parameters, which might cause "overfitting" of the model. Methods of evaluating the accuracy/efficiency of the tradeoffs are proposed.