paper-with-me

홈 › Papers

Likelihood-Free Overcomplete ICA and Applications in Causal Discovery

2019-09-04 · NeurIPS 2019 12 · Chenwei Ding, Mingming Gong, Kun Zhang, DaCheng Tao

Causal discovery witnessed significant progress over the past decades. In particular, many recent causal discovery methods make use of independent, non-Gaussian noise to achieve identifiability of the causal models. Existence of hidden direct common causes, or confounders, generally makes causal discovery more difficult; whenever they are present, the corresponding causal discovery algorithms can be seen as extensions of overcomplete independent component analysis (OICA). However, existing OICA algorithms usually make strong parametric assumptions on the distribution of independent components, which may be violated on real data, leading to sub-optimal or even wrong solutions. In addition, existing OICA algorithms rely on the Expectation Maximization (EM) procedure that requires computationally expensive inference of the posterior distribution of independent components. To tackle these problems, we present a Likelihood-Free Overcomplete ICA algorithm (LFOICA) that estimates the mixing matrix directly by back-propagation without any explicit assumptions on the density function of independent components. Thanks to its computational efficiency, the proposed method makes a number of causal discovery procedures much more practically feasible. For illustrative purposes, we demonstrate the computational efficiency and efficacy of our method in two causal discovery tasks on both synthetic and real data.

📄 PDF Abstract BibTeX arXiv:1909.01525

Code (1)

dingchenwei/Likelihood-free_OICA 공식 구현 pytorch

Tasks

Causal DiscoveryComputational Efficiency

Methods 이 논문이 사용한 방법론

ICA _Independent component analysis (ICA) is a statistical and computational technique for revealing hidden factors that underlie sets of random variables, measurements, or…

Similar Papers 제목 키워드 기반

Causal Discovery of Linear Non-Gaussian Causal Models with Unobserved Confounding

2024-08-09 · Daniela Schkoda, Elina Robeva, Mathias Drton

We consider linear non-Gaussian structural equation models that involve latent confounding. In this setting, the causal structure is identifiable, but, in general, it is not possible to identify the specific causal effec…

Causal Discovery

Training sparse natural image models with a fast Gibbs sampler of an extended state space

2012-12-01 · NeurIPS 2012 12 · Lucas Theis, Jascha Sohl-Dickstein, Matthias Bethge

We present a new learning strategy based on an efficient blocked Gibbs sampler for sparse overcomplete linear models. Particular emphasis is placed on statistical image modeling, where overcomplete models have played an …

Bayesian Model Selection for Identifying Markov Equivalent Causal Graphs

2019-10-16 · pproximateinference AABI Symposium 2019 12 · Mehmet Burak Kurutmaz, Melih Barsbey, Ali Taylan Cemgil, Sinan Yildirim 외

Many approaches to causal discovery are limited by their inability to discriminate between Markov equivalent graphs given only observational data. We formulate causal discovery as a marginal likelihood based Bayesian mod…

Bayesian InferenceCausal DiscoveryModel Selection

Amortized Causal Discovery with Prior-Fitted Networks

2025-12-03 · Mateusz Sypniewski, Mateusz Olko, Mateusz Gajewski, Piotr Miłoś arxiv

In recent years, differentiable penalized likelihood methods have gained popularity, optimizing the causal structure by maximizing its likelihood with respect to the data. However, recent research has shown that errors i…

Empirical Bayesian Approaches for Robust Constraint-based Causal Discovery under Insufficient Data

2022-06-16 · Zijun Cui, Naiyu Yin, Yuru Wang, Qiang Ji

Causal discovery is to learn cause-effect relationships among variables given observational data and is important for many applications. Existing causal discovery methods assume data sufficiency, which may not be the cas…

Causal Discovery