TY - GEN
T1 - Partially linear additive Gaussian graphical models
AU - Geng, Sinong
AU - Yan, Minhao
AU - Kolar, Mladen
AU - Koyejo, Oluwasanmi
N1 - Publisher Copyright:
Copyright © 2019 ASME
PY - 2019
Y1 - 2019
N2 - We propose a partially linear additive Gaussian graphical model (PLA-GGM) for the estimation of associations between random variables distorted by observed confounders. Model parameters are estimated using an Li-regularized maximal pseudo-profile likelihood estimator (MaP-PLE) for which we prove √n-sparsistency. Importantly, our approach avoids parametric constraints on the effects of confounders on the estimated graphical model structure. Empirically, the PLA-GGM is applied to both synthetic and real-world datasets, demonstrating superior performance compared to competing methods.
AB - We propose a partially linear additive Gaussian graphical model (PLA-GGM) for the estimation of associations between random variables distorted by observed confounders. Model parameters are estimated using an Li-regularized maximal pseudo-profile likelihood estimator (MaP-PLE) for which we prove √n-sparsistency. Importantly, our approach avoids parametric constraints on the effects of confounders on the estimated graphical model structure. Empirically, the PLA-GGM is applied to both synthetic and real-world datasets, demonstrating superior performance compared to competing methods.
UR - http://www.scopus.com/inward/record.url?scp=85073236368&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85073236368&partnerID=8YFLogxK
M3 - Conference contribution
AN - SCOPUS:85073236368
T3 - 36th International Conference on Machine Learning, ICML 2019
SP - 3827
EP - 3847
BT - 36th International Conference on Machine Learning, ICML 2019
PB - International Machine Learning Society (IMLS)
T2 - 36th International Conference on Machine Learning, ICML 2019
Y2 - 9 June 2019 through 15 June 2019
ER -