Trading-Off Cost of Deployment Versus Accuracy in Learning Predictive Models

TL;DR

This paper introduces a novel cost-sensitive regularizer framework for predictive models, especially in healthcare, balancing deployment costs with accuracy by leveraging boolean circuit representations of complex cost structures.

Contribution

It presents a new regularizer design based on boolean circuit properties to handle complex cost dependencies in predictive modeling.

Findings

Improved accuracy-cost tradeoff in healthcare risk prediction

Regularizer effectively captures complex cost structures

Models align with real-world deployment cost considerations

Abstract

Predictive models are finding an increasing number of applications in many industries. As a result, a practical means for trading-off the cost of deploying a model versus its effectiveness is needed. Our work is motivated by risk prediction problems in healthcare. Cost-structures in domains such as healthcare are quite complex, posing a significant challenge to existing approaches. We propose a novel framework for designing cost-sensitive structured regularizers that is suitable for problems with complex cost dependencies. We draw upon a surprising connection to boolean circuits. In particular, we represent the problem costs as a multi-layer boolean circuit, and then use properties of boolean circuits to define an extended feature vector and a group regularizer that exactly captures the underlying cost structure. The resulting regularizer may then be combined with a fidelity function to perform model prediction, for example. For the challenging real-world application of risk prediction for sepsis in intensive care units, the use of our regularizer leads to models that are in harmony with the underlying cost structure and thus provide an excellent prediction accuracy versus cost tradeoff.