arXiv 21 Nov 2023 · Machine Learning · 2 citations (OpenAlex)
arXiv:2311.12267 · PDF · DOI · OpenAlex · Extracted main text
We study causal representation learning, the task of recovering high-level latent variables and their causal relationships in the form of a causal graph from low-level observed data (such as text and images), assuming access to observations generated from multiple environments. Prior results on the identifiability of causal representations typically assume access to single-node interventions which is rather unrealistic in practice, since the latent variables are unknown in the first place. In this work, we provide the first identifiability results based on data that stem from general environments. We show that for linear causal models, while the causal graph can be fully recovered, the latent variables are only identified up to the surrounded-node ambiguity (SNA) \citep{varici2023score}. We provide a counterpart of our guarantee, showing that SNA is basically unavoidable in our setting. We also propose an algorithm, LiNGCReL which provably recovers the ground-truth model up to SNA, and we demonstrate its effectiveness via numerical experiments. Finally, we consider general non-parametric causal models and show that the same identification barrier holds when assuming access to groups of soft single-node interventions.
appendix boundary found by appendix_command · 28% of the source is main text. Read the extracted text to check this.
The works this paper leans on most, across its whole bibliography — not restricted to papers in our corpus. Ranked by composite intensity, which combines how often a work is mentioned, how many sections mention it, and how much of that falls in the main text rather than the appendix.
| Reference | Intensity | Mentions | Sections | Main text | |
|---|---|---|---|---|---|
| 1 | Varici, Burak, Acarturk, Emre, Shanmugam, Karthikeyan, Kumar, Abhish… (2023) Score-based causal representation learning with interventions | 0.941 | 6 | 5 | 83% |
| 2 | Kügelgen, Julius, Besserve, Michel, Liang, Wendong, Gresele, Luigi,… (2023) Nonparametric Identifiability of Causal Representations from Unknown Interventions | 0.894 | 7 | 5 | 71% |
| 3 | Acartürk, Emre, Shanmugam, Karthikeyan, Tajer, Ali (2023) General Identifiability and Achievability for Causal Representation Learning | 0.843 | 5 | 5 | 60% |
| 4 | Seigal, Anna, Squires, Chandler, Uhler, Caroline (2022) Linear causal disentanglement via interventions | 0.822 | 9 | 6 | 56% |
| 5 | Zhang, Jiaqi, Squires, Chandler, Greenewald, Kristjan, Srivastava, A… (2023) Identifiability Guarantees for Causal Disentanglement from Soft Interventions | 0.737 | 4 | 3 | 50% |
| 6 | Liang, Wendong, Kekić, Armin, Kügelgen, Julius, Buchholz, Simon, Bes… (2023) Causal Component Analysis | 0.737 | 3 | 3 | 67% |
| 7 | Buchholz, Simon, Rajendran, Goutham, Rosenfeld, Elan, Aragam, Bryon,… (2023) Learning Linear Causal Representations from Interventions under General Nonlinear Mixing | 0.644 | 3 | 2 | 67% |
| 8 | Shimizu, Shohei, Hoyer, Patrik O, Hyvärinen, Aapo, Kerminen, Antti,… (2006) A linear non-Gaussian acyclic model for causal discovery. | 0.644 | 2 | 2 | 100% |
| 9 | Silva, Ricardo, Scheines, Richard, Glymour, Clark, Spirtes, Peter, C… (2006) Learning the Structure of Linear Latent Variable Models. | 0.644 | 2 | 2 | 100% |
| 10 | Tejada-Lapuerta, Alejandro, Bertin, Paul, Bauer, Stefan, Aliee, Hana… (2023) Causal machine learning for single-cell genomics | 0.644 | 2 | 2 | 100% |
Showing the top 10 of 54 scored citations.