@inproceedings{ravanbakhsh_nmr,

  author    = {Ravanbakhsh, Siamak and Poczos, Barnabas and Greiner Russ},
  title     = {A Cross Entropy Optimization Method for Partially Decomposable Problems},
  booktitle = {Proceedings of the Twenty-Fourth AAAI Conference on Artificial Intelligence. Special Track on AI and Bioinformatics},
  pages     = {1280--1286},
  year      = {2010},
  editor    = {M. Fox, D. Poole},
  publisher = {AAAI Press},
  month     = {July 11 {--} 15 },
  address   = {Atlanta, USA},
  order = {100},  
  keywords = {bioinformatics, application},
  url_Paper = {http://www.aaai.org/ocs/index.php/AAAI/AAAI10/paper/view/1848/2196},
  url_Slides = {http://www.cs.cmu.edu/~mravanba/presentations/AAAI_ceed_presentation.pdf},
  abstract = {"Some real-world problems are partially decomposable, in
that they can be decomposed into a set of coupled subproblems,
that are each relatively easy to solve. However,
when these sub-problem share some common variables, it
is not sufficient to simply solve each sub-problem in isolation.
We develop a technology for such problems, and use
it to address the challenge of finding the concentrations of
the chemicals that appear in a complex mixture, based on
its one-dimensional 1H Nuclear Magnetic Resonance (NMR)
spectrum. As each chemical involves clusters of spatially localized
peaks, this requires finding the shifts for the clusters
and the concentrations of the chemicals, that collectively produce
the best match to the observed NMR spectrum. Here,
each sub-problem requires finding the chemical concentrations
and cluster shifts that can appear within a limited spectrum
range; these are coupled as these limited regions can
share many chemicals, and so must agree on the concentrations
and cluster shifts of the common chemicals. This
task motivates CEED: a novel extension to the Cross-Entropy
stochastic optimization method constructed to address such
partially decomposable problems. Our experimental results
in the NMR task show that our CEED system is superior to
other well-known optimization methods, and indeed produces
the best-known results in this important, real-world application."}
}

{"_id":"fqrWfTGMG9vZA9YZm","bibbaseid":"ravanbakhsh-poczos-russ-acrossentropyoptimizationmethodforpartiallydecomposableproblems-2010","downloads":0,"creationDate":"2016-11-13T01:41:56.155Z","title":"A Cross Entropy Optimization Method for Partially Decomposable Problems","author_short":["Ravanbakhsh, S.","Poczos, B.","Russ, G."],"year":2010,"bibtype":"inproceedings","biburl":"http://cs.cmu.edu/~mravanba/conference.bib","bibdata":{"bibtype":"inproceedings","type":"inproceedings","author":[{"propositions":[],"lastnames":["Ravanbakhsh"],"firstnames":["Siamak"],"suffixes":[]},{"propositions":[],"lastnames":["Poczos"],"firstnames":["Barnabas"],"suffixes":[]},{"firstnames":["Greiner"],"propositions":[],"lastnames":["Russ"],"suffixes":[]}],"title":"A Cross Entropy Optimization Method for Partially Decomposable Problems","booktitle":"Proceedings of the Twenty-Fourth AAAI Conference on Artificial Intelligence. Special Track on AI and Bioinformatics","pages":"1280--1286","year":"2010","editor":[{"propositions":[],"lastnames":["M.","Fox"],"firstnames":["D.","Poole"],"suffixes":[]}],"publisher":"AAAI Press","month":"July 11 -- 15 ","address":"Atlanta, USA","order":"100","keywords":"bioinformatics, application","url_paper":"http://www.aaai.org/ocs/index.php/AAAI/AAAI10/paper/view/1848/2196","url_slides":"http://www.cs.cmu.edu/~mravanba/presentations/AAAI_ceed_presentation.pdf","abstract":"\"Some real-world problems are partially decomposable, in that they can be decomposed into a set of coupled subproblems, that are each relatively easy to solve. However, when these sub-problem share some common variables, it is not sufficient to simply solve each sub-problem in isolation. We develop a technology for such problems, and use it to address the challenge of finding the concentrations of the chemicals that appear in a complex mixture, based on its one-dimensional 1H Nuclear Magnetic Resonance (NMR) spectrum. As each chemical involves clusters of spatially localized peaks, this requires finding the shifts for the clusters and the concentrations of the chemicals, that collectively produce the best match to the observed NMR spectrum. Here, each sub-problem requires finding the chemical concentrations and cluster shifts that can appear within a limited spectrum range; these are coupled as these limited regions can share many chemicals, and so must agree on the concentrations and cluster shifts of the common chemicals. This task motivates CEED: a novel extension to the Cross-Entropy stochastic optimization method constructed to address such partially decomposable problems. Our experimental results in the NMR task show that our CEED system is superior to other well-known optimization methods, and indeed produces the best-known results in this important, real-world application.\"","bibtex":"@inproceedings{ravanbakhsh_nmr,\n\n author = {Ravanbakhsh, Siamak and Poczos, Barnabas and Greiner Russ},\n title = {A Cross Entropy Optimization Method for Partially Decomposable Problems},\n booktitle = {Proceedings of the Twenty-Fourth AAAI Conference on Artificial Intelligence. Special Track on AI and Bioinformatics},\n pages = {1280--1286},\n year = {2010},\n editor = {M. Fox, D. Poole},\n publisher = {AAAI Press},\n month = {July 11 {--} 15 },\n address = {Atlanta, USA},\n order = {100}, \n keywords = {bioinformatics, application},\n url_Paper = {http://www.aaai.org/ocs/index.php/AAAI/AAAI10/paper/view/1848/2196},\n url_Slides = {http://www.cs.cmu.edu/~mravanba/presentations/AAAI_ceed_presentation.pdf},\n abstract = {\"Some real-world problems are partially decomposable, in\nthat they can be decomposed into a set of coupled subproblems,\nthat are each relatively easy to solve. However,\nwhen these sub-problem share some common variables, it\nis not sufficient to simply solve each sub-problem in isolation.\nWe develop a technology for such problems, and use\nit to address the challenge of finding the concentrations of\nthe chemicals that appear in a complex mixture, based on\nits one-dimensional 1H Nuclear Magnetic Resonance (NMR)\nspectrum. As each chemical involves clusters of spatially localized\npeaks, this requires finding the shifts for the clusters\nand the concentrations of the chemicals, that collectively produce\nthe best match to the observed NMR spectrum. Here,\neach sub-problem requires finding the chemical concentrations\nand cluster shifts that can appear within a limited spectrum\nrange; these are coupled as these limited regions can\nshare many chemicals, and so must agree on the concentrations\nand cluster shifts of the common chemicals. This\ntask motivates CEED: a novel extension to the Cross-Entropy\nstochastic optimization method constructed to address such\npartially decomposable problems. Our experimental results\nin the NMR task show that our CEED system is superior to\nother well-known optimization methods, and indeed produces\nthe best-known results in this important, real-world application.\"}\n}\n\n\n","author_short":["Ravanbakhsh, S.","Poczos, B.","Russ, G."],"editor_short":["M. Fox, D. P."],"key":"ravanbakhsh_nmr","id":"ravanbakhsh_nmr","bibbaseid":"ravanbakhsh-poczos-russ-acrossentropyoptimizationmethodforpartiallydecomposableproblems-2010","role":"author","urls":{" paper":"http://www.aaai.org/ocs/index.php/AAAI/AAAI10/paper/view/1848/2196"," slides":"http://www.cs.cmu.edu/~mravanba/presentations/AAAI_ceed_presentation.pdf"},"keyword":["bioinformatics","application"],"downloads":0,"html":""},"search_terms":["cross","entropy","optimization","method","partially","decomposable","problems","ravanbakhsh","poczos","russ"],"keywords":["bioinformatics","application"],"authorIDs":["5825f39c4d1f1ec618000052"],"dataSources":["4pQuQhD3X42c59c7K"]}