Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
134 tokens/sec
GPT-4o
10 tokens/sec
Gemini 2.5 Pro Pro
47 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

Multi-granularity Causal Structure Learning (2312.05549v2)

Published 9 Dec 2023 in cs.LG and stat.ML

Abstract: Unveil, model, and comprehend the causal mechanisms underpinning natural phenomena stand as fundamental endeavors across myriad scientific disciplines. Meanwhile, new knowledge emerges when discovering causal relationships from data. Existing causal learning algorithms predominantly focus on the isolated effects of variables, overlook the intricate interplay of multiple variables and their collective behavioral patterns. Furthermore, the ubiquity of high-dimensional data exacts a substantial temporal cost for causal algorithms. In this paper, we develop a novel method called MgCSL (Multi-granularity Causal Structure Learning), which first leverages sparse auto-encoder to explore coarse-graining strategies and causal abstractions from micro-variables to macro-ones. MgCSL then takes multi-granularity variables as inputs to train multilayer perceptrons and to delve the causality between variables. To enhance the efficacy on high-dimensional data, MgCSL introduces a simplified acyclicity constraint to adeptly search the directed acyclic graph among variables. Experimental results show that MgCSL outperforms competitive baselines, and finds out explainable causal connections on fMRI datasets.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (41)
  1. The hippocampus and memory: insights from spatial processing. Nat. Rev. Neurosci., 9(3): 182–194.
  2. A limited memory algorithm for bound constrained optimization. SIAM J Sci. Comput., 16(5): 1190–1208.
  3. Chickering, D. M. 2002. Optimal structure identification with greedy search. JMLR, 3(11): 507–554.
  4. Order-independent constraint-based causal structure learning. JMLR, 15(116): 3921–3962.
  5. Demšar, J. 2006. Statistical comparisons of classifiers over multiple data sets. JMLR, 7: 1–30.
  6. Elder-Vass, D. 2012. Top-down causation and social structures. Interface focus, 2(1): 82–90.
  7. Characterization and greedy learning of interventional Markov equivalence classes of directed acyclic graphs. JMLR, 13(79): 2409–2464.
  8. Daring: Differentiable causal discovery with residual independence. In KDD, 596–605.
  9. Nonlinear causal discovery with additive noise models. In NeuIPS, 689–696.
  10. Gradient-based neural DAG learning. In ICLR.
  11. Scaling structural learning with NO-BEARS to infer causal transcriptome networks. In PSB, 391–402. World Scientific.
  12. Gradient-based local causal structure learning. IEEE TCYB, 99(1): 1–10.
  13. Directed acyclic graph learning on attributed heterogeneous network. TKDE, 99(1): 1–12.
  14. Loewer, B. 2012. The emergence of time’s arrows and special science laws from physics. Interface focus, 2(1): 13–19.
  15. Bayesian network structural learning from complex survey data: a resampling based approach. SMA, 31(4): 981–1013.
  16. Ng, A.; et al. 2011. Sparse autoencoder. CS294A Lecture notes, 72(2011): 1–19.
  17. A graph autoencoder approach to causal structure learning. In NeurIPS Workshop.
  18. Noble, D. 2012. A theory of biological relativity: no privileged level of causation. Interface focus, 2(1): 55–64.
  19. Causal discovery with continuous additive noise models. JMLR, 15: 2009–2053.
  20. Long-term neural and physiological phenotyping of a single human. Nat. Commun., 6(1): 8885.
  21. Meta attention-generation network for cross-granularity few-shot learning. IJCV, 131(5): 1211–1233.
  22. A million variables and more: the fast greedy equivalence search algorithm for learning high-dimensional graphical causal models, with an application to functional magnetic resonance images. JDSA, 3(2): 121–129.
  23. Improving the accuracy of medical diagnosis with causal machine learning. Nat. Commun., 11(1): 3923.
  24. Causal protein-signaling networks derived from multiparameter single-cell data. Science, 308(5721): 523–529.
  25. Toward causal representation learning. Proceedings of the IEEE, 109(5): 612–634.
  26. Causation, prediction, and search. Cambridge, MA: MIT Press, 2nd ed. edition.
  27. Causality and causal inference in epidemiology: the need for a pluralistic approach. Int. J. Epidemiol., 45(6): 1776–1786.
  28. Multivariate information theory uncovers synergistic subsystems of the human cerebral cortex. Commun. Biol., 6(1): 451.
  29. On the fairness of causal algorithmic recourse. In AAAI, 9584–9594.
  30. Wang, G. 2017. DGCC: data-driven granular cognitive computing. Granul. Comput., 2(4): 343–355.
  31. A review of research on multi-granularity cognition based intelligent computing. Chin. J. Comput., 45(6): 1161–1175.
  32. Ordering-based causal discovery with reinforcement learning. In IJCAI, 3566–3573.
  33. Causal inference for recommender systems. In RecSys, 426–431.
  34. End-to-end multi-granulation causality extraction model. DCN.
  35. A multi-granularity network for emotion-cause pair extraction via matrix capsule. In CIKM, 4625–4629.
  36. Reinforcement causal structure learning on order graph. In AAAI, 10737–10744.
  37. DAG-GNN: DAG structure learning with graph neural networks. In ICML, 7154–7163.
  38. Causal discovery from nonstationary/heterogeneous data: skeleton estimation and orientation determination. In IJCAI, 1347–1353.
  39. On the identifiability of the post-nonlinear causal model. In UAI, 647–655.
  40. Dags with no tears: continuous optimization for structure learning. In NeuIPS, 9492–9503.
  41. Learning sparse nonparametric dags. In AISTATS, 3414–3425.

Summary

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

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