Least-Squares Independence Regression for Non-Linear Causal Inference under Non-Gaussian Noise

Details Least-Squares Independence Regression for Non-Linear Causal Inference under Non-Gaussian Noise Least-Squares Independence Regression for Non-Linear Causal Inference under Non-Gaussian Noise

2011-03-29

arxiv.org/abs/1103.5537v2

Statistics

Machine Learning

submitted to Machine Learning Journal

Scientific paper

The discovery of non-linear causal relationship under additive non-Gaussian noise models has attracted considerable attention recently because of their high flexibility. In this paper, we propose a novel causal inference algorithm called least-squares independence regression (LSIR). LSIR learns the additive noise model through the minimization of an estimator of the squared-loss mutual information between inputs and residuals. A notable advantage of LSIR over existing approaches is that tuning parameters such as the kernel width and the regularization parameter can be naturally optimized by cross-validation, allowing us to avoid overfitting in a data-dependent fashion. Through experiments with real-world datasets, we show that LSIR compares favorably with a state-of-the-art causal inference method.

Affiliated with

Also associated with

No associations

Rating

Least-Squares Independence Regression for Non-Linear Causal Inference under Non-Gaussian Noise does not yet have a rating. At this time, there are no reviews or comments for this scientific paper.

If you have personal experience with Least-Squares Independence Regression for Non-Linear Causal Inference under Non-Gaussian Noise, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Least-Squares Independence Regression for Non-Linear Causal Inference under Non-Gaussian Noise will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFWR-SCP-O-162641