@article{lengerich2022notmad,
  title={NOTMAD: Estimating Bayesian Networks with Sample-Specific Structures and Parameters},
  author={Lengerich, Benjamin and Ellington, Caleb and Aragam, Bryon and Xing, Eric P. and Kellis, Manolis},
  informal_venue = {arXiv},
  url_preprint={https://arxiv.org/abs/2111.01104},
  abstract={Context-specific Bayesian networks (i.e. directed acyclic graphs, DAGs) identify context-dependent relationships between variables, but the non-convexity induced by the acyclicity requirement makes it difficult to share information between context-specific estimators (e.g. with graph generator functions). For this reason, existing methods for inferring context-specific Bayesian networks have favored breaking datasets into subsamples, limiting statistical power and resolution, and preventing the use of multidimensional and latent contexts. To overcome this challenge, we propose NOTEARS-optimized Mixtures of Archetypal DAGs (NOTMAD). NOTMAD models context-specific Bayesian networks as the output of a function which learns to mix archetypal networks according to sample context. The archetypal networks are estimated jointly with the context-specific networks and do not require any prior knowledge. We encode the acyclicity constraint as a smooth regularization loss which is back-propagated to the mixing function; in this way, NOTMAD shares information between context-specific acyclic graphs, enabling the estimation of Bayesian network structures and parameters at even single-sample resolution. We demonstrate the utility of NOTMAD and sample-specific network inference through analysis and experiments, including patient-specific gene expression networks which correspond to morphological variation in cancer.},
  keywords={Bayesian Networks, Contextualized},
  year={2022}
}

{"_id":"LzjEE5jTMnyzAqWf2","bibbaseid":"lengerich-ellington-aragam-xing-kellis-notmadestimatingbayesiannetworkswithsamplespecificstructuresandparameters-2022","author_short":["Lengerich, B.","Ellington, C.","Aragam, B.","Xing, E. P.","Kellis, M."],"bibdata":{"bibtype":"article","type":"article","title":"NOTMAD: Estimating Bayesian Networks with Sample-Specific Structures and Parameters","author":[{"propositions":[],"lastnames":["Lengerich"],"firstnames":["Benjamin"],"suffixes":[]},{"propositions":[],"lastnames":["Ellington"],"firstnames":["Caleb"],"suffixes":[]},{"propositions":[],"lastnames":["Aragam"],"firstnames":["Bryon"],"suffixes":[]},{"propositions":[],"lastnames":["Xing"],"firstnames":["Eric","P."],"suffixes":[]},{"propositions":[],"lastnames":["Kellis"],"firstnames":["Manolis"],"suffixes":[]}],"informal_venue":"arXiv","url_preprint":"https://arxiv.org/abs/2111.01104","abstract":"Context-specific Bayesian networks (i.e. directed acyclic graphs, DAGs) identify context-dependent relationships between variables, but the non-convexity induced by the acyclicity requirement makes it difficult to share information between context-specific estimators (e.g. with graph generator functions). For this reason, existing methods for inferring context-specific Bayesian networks have favored breaking datasets into subsamples, limiting statistical power and resolution, and preventing the use of multidimensional and latent contexts. To overcome this challenge, we propose NOTEARS-optimized Mixtures of Archetypal DAGs (NOTMAD). NOTMAD models context-specific Bayesian networks as the output of a function which learns to mix archetypal networks according to sample context. The archetypal networks are estimated jointly with the context-specific networks and do not require any prior knowledge. We encode the acyclicity constraint as a smooth regularization loss which is back-propagated to the mixing function; in this way, NOTMAD shares information between context-specific acyclic graphs, enabling the estimation of Bayesian network structures and parameters at even single-sample resolution. We demonstrate the utility of NOTMAD and sample-specific network inference through analysis and experiments, including patient-specific gene expression networks which correspond to morphological variation in cancer.","keywords":"Bayesian Networks, Contextualized","year":"2022","bibtex":"@article{lengerich2022notmad,\n title={NOTMAD: Estimating Bayesian Networks with Sample-Specific Structures and Parameters},\n author={Lengerich, Benjamin and Ellington, Caleb and Aragam, Bryon and Xing, Eric P. and Kellis, Manolis},\n informal_venue = {arXiv},\n url_preprint={https://arxiv.org/abs/2111.01104},\n abstract={Context-specific Bayesian networks (i.e. directed acyclic graphs, DAGs) identify context-dependent relationships between variables, but the non-convexity induced by the acyclicity requirement makes it difficult to share information between context-specific estimators (e.g. with graph generator functions). For this reason, existing methods for inferring context-specific Bayesian networks have favored breaking datasets into subsamples, limiting statistical power and resolution, and preventing the use of multidimensional and latent contexts. To overcome this challenge, we propose NOTEARS-optimized Mixtures of Archetypal DAGs (NOTMAD). NOTMAD models context-specific Bayesian networks as the output of a function which learns to mix archetypal networks according to sample context. The archetypal networks are estimated jointly with the context-specific networks and do not require any prior knowledge. We encode the acyclicity constraint as a smooth regularization loss which is back-propagated to the mixing function; in this way, NOTMAD shares information between context-specific acyclic graphs, enabling the estimation of Bayesian network structures and parameters at even single-sample resolution. We demonstrate the utility of NOTMAD and sample-specific network inference through analysis and experiments, including patient-specific gene expression networks which correspond to morphological variation in cancer.},\n keywords={Bayesian Networks, Contextualized},\n year={2022}\n}\n\n","author_short":["Lengerich, B.","Ellington, C.","Aragam, B.","Xing, E. P.","Kellis, M."],"key":"lengerich2022notmad","id":"lengerich2022notmad","bibbaseid":"lengerich-ellington-aragam-xing-kellis-notmadestimatingbayesiannetworkswithsamplespecificstructuresandparameters-2022","role":"author","urls":{" preprint":"https://arxiv.org/abs/2111.01104"},"keyword":["Bayesian Networks","Contextualized"],"metadata":{"authorlinks":{}},"html":""},"bibtype":"article","biburl":"http://web.mit.edu/~blengeri/www/lengerich_website.bib","dataSources":["ezuecqGQYxpz639hf","MYQReohz2EeYDJDva","sRozo3iEee6h28fas"],"keywords":["bayesian networks","contextualized"],"search_terms":["notmad","estimating","bayesian","networks","sample","specific","structures","parameters","lengerich","ellington","aragam","xing","kellis"],"title":"NOTMAD: Estimating Bayesian Networks with Sample-Specific Structures and Parameters","year":2022}