# Invariance, Causality and Robustness

## TL;DR

This paper unifies causal inference and predictive robustness through probabilistic invariance, proposing a new risk minimization approach that enhances robustness and interpretability in machine learning models.

## Contribution

It introduces a novel methodology leveraging invariance for causal inference and robustness, applicable to heterogeneous data, improving interpretability and resilience of predictions.

## Key findings

- Invariance can be estimated from diverse perturbation data.
- The approach offers increased robustness over standard methods.
- Provides causal insights with better interpretability.

## Abstract

We discuss recent work for causal inference and predictive robustness in a unifying way. The key idea relies on a notion of probabilistic invariance or stability: it opens up new insights for formulating causality as a certain risk minimization problem with a corresponding notion of robustness. The invariance itself can be estimated from general heterogeneous or perturbation data which frequently occur with nowadays data collection. The novel methodology is potentially useful in many applications, offering more robustness and better `causal-oriented' interpretation than machine learning or estimation in standard regression or classification frameworks.