Position: The Causal Revolution Needs Scientific Pragmatism

TL;DR

The paper advocates for a pragmatic, human-centric approach to causal modeling in machine learning, emphasizing the need to balance scientific rigor with practical application to advance the causal revolution.

Contribution

It introduces the concept of scientific pragmatism as a guiding philosophy to overcome current limitations and promote the adoption of causal methods in machine learning.

Findings

Current causal models face adoption barriers due to conflicting worldviews.

A shift towards scientific pragmatism can facilitate the integration of causality in ML.

Balancing assumptions and practicality is key to advancing causal methods.

Abstract

Causal models and methods have great promise, but their progress has been stalled. Proposals using causality get squeezed between two opposing worldviews. Scientific perfectionism--an insistence on only using "correct" models--slows the adoption of causal methods in knowledge generating applications. Pushing in the opposite direction, the academic discipline of computer science prefers algorithms with no or few assumptions, and technologies based on automation and scalability are often selected for economic and business applications. We argue that these system-centric inductive biases should be replaced with a human-centric philosophy we refer to as scientific pragmatism. The machine learning community must strike the right balance to make space for the causal revolution to prosper.