_____________________________ NIPS 2018 Workshop on Machine Learning for Health (NIPS ML4H 2018) Moving beyond supervised learning in healthcare
A workshop at the Thirty-Second Annual Conference on Neural Information Processing Systems (NIPS 2018) Saturday, December 8, 2018 Palais des Congrès de Montréal, Montréal Canada https://ml4health.github.io/2018/ Please direct questions to: [email protected] NOTE: Historically the main NIPS conference has sold out quickly, and this may extend to workshop registrations. If you plan to submit a paper, please register as soon as possible. Registration opens Sep. 4, 2018 (https://nips.cc/Register/view-registration) can be cancelled before November 15, 2018, 11:59 pacific time for a full refund (https://nips.cc/Help/CancellationPolicy). DATES: * Mon Oct 29, 2018: Submission deadline at 11:59pm * Mon Nov 12, 2018: Acceptance notification (Poster or Spotlight+Poster) * Fri Nov 15, 2018: NIPS deadline to cancel registration (with full refund) * Fri Nov 30, 2018: Final papers posted online (with permission) * Sat, Dec 8, 2018: Workshop ========================================================================= ABSTRACT: The goal of the Machine Learning for Health Workshop (NIPS ML4H 2018) is to foster collaborations that meaningfully impact medicine by bringing together clinicians, health data experts, and machine learning researchers. We aim to build on the success of the last three NIPS ML4H workshops which were widely attended and helped form the foundations of a new research community. This year’s workshop will focus on fostering innovative new strategies for applying machine learning in healthcare and medicine. To date, nearly all of the most notable successes of machine learning have been driven by supervised learning. In this workshop, we will convene a diverse set of leading researchers that will expose attendees to a broader class of computational solutions to the challenges facing clinical medicine. Speakers will highlight a range of important clinical problems, and focus discussion on opportunities for diverse methods including clustering, active learning, dimensionality reduction, reinforcement learning, and causal inference. The workshop will feature invited talks from leading voices in both medicine and machine learning. Invited clinicians will discuss open clinical problems where data-driven solutions can make an immediate difference. The workshop will conclude with an interactive panel discussion where all speakers respond to questions provided by the audience. From the research community, we welcome short paper submissions highlighting novel research contributions at the intersection of machine learning and healthcare. Accepted submissions will be featured as poster presentations and (in select cases) as short oral spotlight presentations. While our emphasis this year will focus on moving beyond supervised learning, we welcome any innovative submission seeking to use ML to improve healthcare. We encourage submissions from all researchers, regardless of background or prior experience. SUBMISSION INSTRUCTIONS: Researchers interested in contributing should upload short, anonymized papers of up to 4 pages in PDF format by Friday, October 26, 2018, 11:59 PM in the timezone of your choice. The organizers strongly encourage submissions from all, regardless of background and prior work. Please submit via our ML4H EasyChair website: https://easychair.org/conferences/?conf=nipsml4h2018 Papers should adhere to the NIPS conference paper format, via the NIPS LaTeX style file: https://nips.cc/Conferences/2018/PaperInformation/StyleFiles Workshop papers should be at most 4 pages of content, including text and figures. Additional pages containing only bibliographic references can be included without penalty. Authors will not be penalized for including an appendix of supplementary material after the references. However, reviewers will not be required to consult any appendices to make their decisions. The main 4-page paper should adequately describe the work and its contributions. #### Relevant Topics Submitted papers should describe innovative machine learning research focused on relevant problems in health and medicine. This can mean new models, new datasets, new algorithms, or new applications. Topics of interest include but are not limited to reinforcement learning, temporal models, deep learning, semi-supervised learning, data integration, learning from missing or biased data, learning from non-stationary data, model criticism, model interpretability, causality, model biases, and transfer learning. #### Peer Review and Acceptance Criteria In order to ensure submissions are relevant and clear, all submissions must include a one paragraph abstract that is technically precise and 1) identifies the problem, 2) motivates its importance, 3) indicates the methods used, and 4) provides a summary of results. Reviewers will check if the body of work supports the claim made in the abstract. All submissions will undergo double-blind peer review. It will be up to the authors to ensure the proper anonymization of their paper. Do not include any names or affiliations. Refer to your own past work in the third-person. Accepted papers will be chosen based on technical merit and suitability to the workshop's goals. All accepted papers will be included in one of two poster presentation sessions on the day of the workshop. Some accepted papers will be invited to give short oral spotlight presentations at the workshop. #### Registration and Attendance To promote community interaction, at least one presenting author should register for the workshop. Historically the main NIPS conference has sold out quickly, and this may extend to workshop registrations. If you plan to submit a paper, please register as soon as possible. Registration opens Sep. 4, 2018 (https://nips.cc/Register/view-registration) can be cancelled before November 15, 2018, 11:59 pacific time for a full refund (https://nips.cc/Help/CancellationPolicy). If you have already registered, confirm that you have a valid workshop registration. If NIPS workshop registration has sold out, we encourage researchers to submit a paper regardless of their registration status; however, they should notify the organizers in case additional workshop registrations are made available: [email protected]. If your paper is accepted and you cannot attend due to registration or other issues, please contact us after you are accepted and we'll find solutions on a case-by-case basis. Acceptance notifications will go out a few days before the NIPS deadline for full refunds. #### Copyright for Accepted Papers This workshop will be informally published online but not officially archived. This means: Authors can retain full copyright of their papers. Acceptance to NIPS ML4H 2018 does not preclude publication of the same material in another journal or conference. We encourage (but do not require) accepted papers to be posted on arXiv. With author permission, we will post links to accepted short papers on our workshop website. Our workshop does allow submission of papers that are under review or have been recently published in a conference or a journal. Authors should clearly state any overlapping published work at the time of submission. ORGANIZERS: Tristan Naumann (Microsoft Research) Brett Beaulieu-Jones (Harvard Medical School) Samuel Finlayson (Harvard Medical School) Irene Chen (Massachusetts Institute of Technology) Andrew Beam (Harvard Medical School) Marzyeh Ghassemi (Verily, MIT, UToronto, Vector) Matthew McDermott (Massachusetts Institute of Technology) Madalina Fiterau (Stanford, UMass Amherst) Michael Hughes (Tufts University and Harvard) Farah Shamout (Oxford) Corey Chivers (University of Pennsylvania) Jaz Kandola (Imperial College London) Alexandre Yahi (Columbia University) Samuel Finlayson (Harvard Medical School) Bruno Jedynak (Portland State University) Peter Schulam (Johns Hopkins University) Natalia Antropova (University of Chicago) Jason Fries (Stanford) Adrian Dalca (MIT and Harvard Medical School) _______________________________________________ uai mailing list [email protected] https://secure.engr.oregonstate.edu/mailman/listinfo/uai
