BMAW 2015
Annual Bayesian Modeling

Applications Workshop

16 July 2015 - Workshop Day


Amsterdam, The Netherlands


Continuing a successful tradition as part of the UAI conference, the Twelfth Annual Bayesian Modeling Applications Workshop will provide a forum for exchange about real-world problems among applications practitioners, tool developers, and researchers. The aim of the workshop is to foster discussion on the challenges of building applications whilst considering stakeholders, user interaction, tools, knowledge elicitation, learning, validation, system integration, and deployment.

The theme of the Workshop has adapted from year to year, as real-world problems change and technologies evolve to meet them. The frenzy to apply conventional machine learning methods for commercial applications has the danger of overwhelming Bayesian methods where they might be best applied. Bayesian methods face a similar challenge to the one they faced a decade ago by this community: To demonstrate their timeliness in the current environment of intelligent systems and a long tail of related decision and prediction tasks.

Submissions are solicited of real-world applications of Bayesian models and computational methods, to be presented in oral or poster sessions. There will also be opportunity to take part in discussions with invited speakers and with a concluding panel about the state of the current application environment. We encourage submissions from a broad spectrum of topics, with a bias toward novel domains, including those that cross domains or disciplines, on topics suggested by this list:

- Problem representation, formulation, and model design;

- Novel approaches for learning and inference with probabilistic graphical models, inspired by a real-world problem;

- Combining machine learning, active learning, and elicitation of expert knowledge;

- Tractable, scalable computational methods for complex and distributed models;

- Bayesian approaches to classification, clustering, recommendations, personalization, search, advertising, and on-line so-called "A/B" testing;

- Generative models for data cleaning, feature engineering, or fusion of different types of data;

- Techniques for domains with missing, incomplete, large, heterogeneous, and unstructured data;

- Extensions to relational data, causality, infinite domains, spatial or temporal reasoning.