Score-based causal discovery methods

In this section, we would like to introduce Score-based causal discovery methods, including GES and Exact search methods. For GES, we implemented it with BIC score [1] and generalized score [2]. For Exact search, we implemented DP [3] and A* [4].

Contents:

• GES with the BIC score or generalized score
  • Algorithm Introduction
  • Usage
  • Parameters
  • Returns
• Exact Search
  • Algorithm Introduction
  • Usage
  • Parameters
  • Returns

[1]

Chickering, D. M. (2002). Optimal structure identification with greedy search. Journal of machine learning research, 3(Nov), 507-554.

[2]

Huang, B., Zhang, K., Lin, Y., Schölkopf, B., & Glymour, C. (2018, July). Generalized score functions for causal discovery. In Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining (pp. 1551-1560).

[3]

Silander, T., & Myllymäki, P. (2006, July). A simple approach for finding the globally optimal Bayesian network structure. In Proceedings of the Twenty-Second Conference on Uncertainty in Artificial Intelligence (pp. 445-452).

[4]

Yuan, C., & Malone, B. (2013). Learning optimal Bayesian networks: A shortest path perspective. Journal of Artificial Intelligence Research, 48, 23-65.