Towards Causal Representation Learning

Provides a canonical overview or reference point for the relevant DoOperator research area.

A field-defining overview connecting causal inference to representation learning, transfer, modularity, generalization, and discovery of high-level causal variables.

Use when orienting a new paper, blog post, benchmark, or research plan in this area.

Do not cite the overview as evidence that a specific method works in a specific deployment setting without checking the underlying primary paper.