Now showing items 7-9 of 9

    • Semi-supervised Regression with Order Preferences 

      Zhu, Xiaojin; Goldberg, Andrew (University of Wisconsin-Madison Department of Computer Sciences, 2006)
      Following a discussion on the general form of regularization for semi-supervised learning, we propose a semi-supervised regression algorithm. It is based on the assumption that we have certain order preferences on unlabeled ...
    • Some Submodular Data-Poisoning Attacks on Machine Learners 

      Mei, Shike; Zhu, Xiaojin (2017-03-08)
      We study data-poisoning attacks using a machine teaching framework. For a family of NP-hard attack problems we pose them as submodular function maximization, thereby inheriting efficient greedy algorithms with theoretical ...
    • To Join or Not to Join? Thinking Twice about Joins before Feature Selection 

      Kumar, Arun; Naughton, Jeffrey; Patel, Jignesh M.; Zhu, Xiaojin (2015-11-27)
      Closer integration of machine learning (ML) with data processing is a booming area in both the data management industry and academia. Almost all ML toolkits assume that the input is a single table, but many datasets are ...