講演抄録/キーワード |
講演名 |
2010-06-15 15:50
Dependence Minimizing Regression with Model Selection for Non-Linear Causal Inference under Non-Gaussian Noise ○Makoto Yamada・Masashi Sugiyama(Tokyo Inst. of Tech.) IBISML2010-22 |
抄録 |
(和) |
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 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 the state-of-the-art causal inference method. |
(英) |
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 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 the state-of-the-art causal inference method. |
キーワード |
(和) |
Dependence minimizing regression / causal inference / least-squares independence regression / / / / / |
(英) |
Dependence minimizing regression / causal inference / least-squares independence regression / / / / / |
文献情報 |
信学技報, vol. 110, no. 76, IBISML2010-22, pp. 147-153, 2010年6月. |
資料番号 |
IBISML2010-22 |
発行日 |
2010-06-07 (IBISML) |
ISSN |
Print edition: ISSN 0913-5685 Online edition: ISSN 2432-6380 |
著作権に ついて |
技術研究報告に掲載された論文の著作権は電子情報通信学会に帰属します.(許諾番号:10GA0019/12GB0052/13GB0056/17GB0034/18GB0034) |
PDFダウンロード |
IBISML2010-22 |