@conference {london:aistats14, title = {PAC-Bayesian Collective Stability}, booktitle = {Proceedings of the 17th International Conference on Artificial Intelligence and Statistics}, year = {2014}, author = {London, Ben and Huang, Bert and Benjamin Taskar and Lise Getoor} } @conference {london:icml13, title = {Collective Stability in Structured Prediction: Generalization from One Example}, booktitle = {ICML}, year = {2013}, abstract = {

Structured predictors enable joint inference over multiple interdependent output variables. These models are often trained on a small number of examples with large internal structure. Existing distribution-free generalization bounds do not guarantee generalization in this setting, though this contradicts a large body of empirical evidence from computer vision, natural language processing, social networks and other fields. In this paper, we identify a set of natural conditions {\textendash} weak dependence, hypothesis complexity and a new measure, collective stability {\textendash} that are sufficient for generalization from even a single example, without imposing an explicit generative model of the data. We then demonstrate that the complexity and stability conditions are satisfied by a broad class of models, including marginal inference in templated graphical models. We thus obtain uniform convergence rates that can decrease significantly faster than previous bounds, particularly when each structured example is sufficiently large and the number of training examples is constant, even one.

}, author = {Ben London and Bert Huang and Benjamin Taskar and Lise Getoor} } @conference {huang:slg13, title = {Empirical Analysis of Collective Stability}, booktitle = {ICML Workshop on SLG}, year = {2013}, abstract = {

When learning structured predictors, collective stability is an important factor for generalization. London et al. (2013) provide the first analysis of this effect, proving that collectively stable hypotheses produce less deviation between empirical risk and true risk, i.e., defect. We test this effect empirically using a collectively stable variant of maxmargin Markov networks. Our experiments on webpage classification validate that increasing the collective stability reduces the defect and can thus lead to lower overall test error

}, author = {Bert Huang and Ben London and Benjamin Taskar and Lise Getoor} } @conference {london:nips13ws, title = {PAC-Bayes Generalization Bounds for Randomized Structured Prediction}, booktitle = {NIP Workshop on Perturbation, Optimization and Statistics}, year = {2013}, author = {London, Ben and Huang, Bert and Benjamin Taskar and Lise Getoor} } @book {koller:gm-ch-srl-book07, title = {Graphical Models in a Nutshell}, series = {An Introduction to Statistical Relational Learning}, volume = {1}, year = {2007}, pages = {13--55}, publisher = {MIT Press}, organization = {MIT Press}, edition = {1}, chapter = {2}, abstract = {

Probabilistic graphical models are an elegant framework which combines uncertainty (probabilities) and logical structure (independence constraints) to compactly represent complex, real-world phenomena. The framework is quite general in that many of the commonly proposed statistical models (Kalman filters, hidden Markov models, Ising models) can be described as graphical models. Graphical models have enjoyed a surge of interest in the last two decades, due both to the flexibility and power of the representation and to the increased ability to effectively learn and perform inference in large networks.

}, author = {Daphne Koller and Nir Friedman and Lise Getoor and Benjamin Taskar}, editor = {Lise Getoor and Benjamin Taskar} } @book {getoor:srlbook07, title = {Introduction to Statistical Relational Learning}, year = {2007}, publisher = {The MIT Press}, organization = {The MIT Press}, author = {Lise Getoor and Benjamin Taskar} } @book {getoor:prm-ch-srl-book07, title = {Probabilistic Relational Models}, series = {An Introduction to Statistical Relational Learning}, volume = {1}, year = {2007}, pages = {129--174}, publisher = {MIT Press}, organization = {MIT Press}, edition = {1}, chapter = {5}, abstract = {

Probabilistic relational models (PRMs) are a rich representation language for structured statistical models. They combine a frame-based logical representation with probabilistic semantics based on directed graphical models (Bayesian networks). This chapter gives an introduction to probabilistic relational models, describing semantics for attribute uncertainty, structural uncertainty, and class uncertainty. For each case, learning algorithms and some sample results are presented.

}, author = {Lise Getoor and Nir Friedman and Daphne Koller and Avi Pfeffer and Benjamin Taskar}, editor = {Lise Getoor and Benjamin Taskar} } @article {getoor:jmlr02, title = {Learning Probabilistic Models of Link Structure}, journal = {Journal of Machine Learning Research}, volume = {3}, year = {2002}, pages = {679- -707}, author = {Lise Getoor and Friedman, Nir and Koller, Daphne and Benjamin Taskar} } @conference {getoor:icml01, title = {Learning Probabilistic Models of Relational Structure}, booktitle = {Proceedings of International Conference on Machine Learning (ICML)}, year = {2001}, author = {Lise Getoor and Friedman, Nir and Koller, Daphne and Benjamin Taskar} } @conference {getoor:ijcaiws01, title = {Probabilistic Models of Text and Link Structure for Hypertext Classification}, booktitle = {IJCAI Workshop on Text Learning: Beyond Supervision}, year = {2001}, author = {Lise Getoor and Segal, Eran and Benjamin Taskar and Koller, Daphne} } @conference {getoor:sigmod01, title = {Selectivity estimation using probabilistic relational models}, booktitle = {Proceedings of ACM-SIGMOD 2001 International Conference on Management of Data}, year = {2001}, author = {Lise Getoor and Koller, Daphne and Benjamin Taskar} } @conference {getoor:srl00, title = {Learning Probabilistic Relational Models with Structural Uncertainty}, booktitle = {Proceedings of the AAAI Workshop on Learning Statistical Models from Relational Data}, year = {2000}, author = {Lise Getoor and Koller, Daphne and Benjamin Taskar and Friedman, Nir} }