Goals and tasks

• RDBMS-based grounding (maybe clear, belongs to cutting plane algorithm point)
• Advanced Message Passing algorithms (AD3, dual decomposition, MPLP, column generation etc.)
• Incorporating off-the-shelf (I)LP solvers
• Structural / Combinatorial Factors (e.g. tree constraint)
• Compilation Strategies
• Implementing first order lazy/cutting plane algorithms
• Sampling Methods
• Inference and Learning with Expectation Propagation?
• Structure Learning
• Interactive interface (a la IPython notebook etc.), Interactive tutorial (like here)
• Continuous Benchmarking
• Distributed Cloud-Based service?
• Joint relational tensor/matrix factorization and upstream structured prediction (this is actually the one thing I get the money for that pays this so it would be great if we could think about it)
• Auto-encoders?
• ...

Of course not all of these will be feasible during the codefest.

Applications, Datasets, Models

• Standard MLN examples from here here. For inference, we could use the already prepared ones from here
• Ontology Matching (on Jans Computer ;-)).
• Ontology Learning (also on Jans Computer).
• UCI [datasets] (http://archive.ics.uci.edu/ml/datasets.html).
• Bio corpora for NER, NEN and RE from here
• research-quality data sets