Fast-and-Frugal Trees

Fast-and-frugal trees (FFTs) are a type of decision tree designed for quick, efficient decision-making by using a minimal amount of information and a simple set of rules. Unlike traditional complex decision trees, FFTs aim to make good-enough decisions by considering only a few key attributes, thus embodying the principle of bounded rationality where decision-makers have limited time, information, and computational resources.

Each node in a fast-and-frugal tree examines only one attribute, leading to a decision or to another attribute, until a conclusion is reached. This approach contrasts with more complex models that may consider many attributes simultaneously and require extensive computation. Fast-and-frugal trees are particularly useful in situations where decisions must be made quickly, under uncertainty, or with limited data availability.

‍

In emergency medicine, a fast-and-frugal tree might help triage patients based on a few critical symptoms or signs. For instance, a FFT could be designed to quickly assess whether a patient with chest pain is at high risk of a heart attack, using just a few key indicators such as pain severity, location, and patient's age and medical history. The tree would guide the medical professional through a series of yes/no decisions, leading to a rapid initial assessment and decision on the need for immediate intervention.

Another example is in finance, where a FFT could assist in making quick investment decisions based on a few crucial market indicators, such as sudden changes in stock prices, trading volume, and economic news, to decide whether to buy, hold, or sell a stock. This approach enables investors to react swiftly to market changes without the need for complex analyses.

‍