Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
144 tokens/sec
GPT-4o
7 tokens/sec
Gemini 2.5 Pro Pro
45 tokens/sec
o3 Pro
4 tokens/sec
GPT-4.1 Pro
38 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

Identification for Tree-shaped Structural Causal Models in Polynomial Time (2311.14058v2)

Published 23 Nov 2023 in cs.AI and cs.DS

Abstract: Linear structural causal models (SCMs) are used to express and analyse the relationships between random variables. Direct causal effects are represented as directed edges and confounding factors as bidirected edges. Identifying the causal parameters from correlations between the nodes is an open problem in artificial intelligence. In this paper, we study SCMs whose directed component forms a tree. Van der Zander et al. (AISTATS'22, PLMR 151, pp. 6770--6792, 2022) give a PSPACE-algorithm for the identification problem in this case, which is a significant improvement over the general Gr\"obner basis approach, which has doubly-exponential time complexity in the number of structural parameters. In this work, we present a randomized polynomial-time algorithm, which solves the identification problem for tree-shaped SCMs. For every structural parameter, our algorithms decides whether it is generically identifiable, generically 2-identifiable, or generically unidentifiable. (No other cases can occur.) In the first two cases, it provides one or two fractional affine square root terms of polynomials (FASTPs) for the corresponding parameter, respectively.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (27)
  1. Bollen, K. A. 1989. Structural equations with latent variables. Wiley.
  2. Instrumental Variables. Cambridge University Press.
  3. Brito, C. 2010. Instrumental sets. In Dechter, R.; Geffner, H.; and Halpern, J. Y., eds., Heuristics, Probability and Causality. A Tribute to Judea Pearl, chapter 17, 295–308. College Publications.
  4. Generalized Instrumental Variables. In Proc. Conference on Uncertainty in Artificial Intelligence (UAI), 85–93.
  5. A graphical criterion for the identification of causal effects in linear models. In Proc. AAAI Conference on Artificial Intelligence (AAAI), 533–538.
  6. Incorporating knowledge into structural equation models using auxiliary variables. In Proc. Int. Joint Conference on Artificial Intelligence (IJCAI), 3577–3583.
  7. Ideals, Varieties, and Algorithms. Springer.
  8. Drton, M. 2018. Algebraic problems in structural equation modeling. In The 50th anniversary of Gröbner bases, 35–86. Mathematical Society of Japan.
  9. Duncan, O. D. 1975. Introduction to structural equation models. Academic Press.
  10. Half-trek criterion for generic identifiability of linear structural equation models. The Annals of Statistics, 40(3): 1682–1713.
  11. Identifying Causal Effects with Computer Algebra. In Proc. Conference on Uncertainty in Artificial Intelligence (UAI), 193–200.
  12. Discriminants, resultants, and multidimensional determinants. Birkhäuser.
  13. Krantz, S. G. 1999. Handbook of Complex Variables, chapter Linear Fractional Transformations. Birkhäuser.
  14. Efficient Identification in Linear Structural Causal Models with Instrumental Cutsets. In Proc. Advances in Neural Information Processing Systems (NeurIPS), 12477–12486.
  15. Efficient identification in linear structural causal models with auxiliary cutsets. In Proc. Int. Conference on Machine Learning (ICML), 5501–5510.
  16. Pearl, J. 2001. Parameter identification: A new perspective. Technical Report R-276, UCLA.
  17. Pearl, J. 2009. Causality. Cambridge University Press.
  18. Saptharishi, R.; et al. 2021. A selection of lower bounds in arithmetic circuit complexity. Version 2021-07-27.
  19. Schwartz, J. T. 1980. Fast probabilistic algorithms for verification of polynomial identities. Journal of the ACM, 27(4): 701–717.
  20. Stanley, R. P. 2018. Algebraic Combinatorics. Springer.
  21. On Searching for Generalized Instrumental Variables. In Proc. International Conference on Artificial Intelligence and Statistics (AISTATS), 1214–1222.
  22. Efficiently Finding Conditional Instruments for Causal Inference. In Proc. Int. Joint Conference on Artificial Intelligence (IJCAI), 3243–3249.
  23. Identification in Tree-shaped Linear Structural Causal Models. In Camps-Valls, G.; Ruiz, F. J. R.; and Valera, I., eds., International Conference on Artificial Intelligence and Statistics, AISTATS 2022, 28-30 March 2022, Virtual Event, volume 151 of Proceedings of Machine Learning Research, 6770–6792. PMLR.
  24. Determinantal generalizations of instrumental variables. Journal of Causal Inference, 6(1).
  25. Wright, S. 1921. Correlation and causation. J. Agricultural Research, 20: 557–585.
  26. Wright, S. 1934. The method of path coefficients. The Annals of Mathematical Statistics, 5(3): 161–215.
  27. Zippel, R. 1979. Probabilistic algorithms for sparse polynomials. Symbolic and algebraic computation, 216–226.
Citations (1)
View on Semantic Scholar

Summary

We haven't generated a summary for this paper yet.

Ai Sparkles Streamline Icon: https://streamlinehq.com Summarize Now