As part of the Need for Speed: Emerging Applications for Parallel Computing seminar series, Dan Roth, a professor of computer science at the University of Illinois at Urbana-Champaign, will present an online lecture entitled "Learning and Inference in Natural Language Understanding" on Wednesday, February 25th at 4:15 PM CT. The lecture will be delivered live via video streaming.

Making decisions in natural language understanding tasks often involves assigning values to sets of interdependent variables where an expressive dependency structure among these can influence, or even dictate, what assignments are possible. Structured learning problems provide one such example, but we are interested in a broader setting where multiple models are involved and it may not be ideal, or possible, to learn them jointly.

Professor Roth will present work on Constrained Conditional Models (CCMs), a framework that augments probabilistic models with declarative constraints as a way to support decisions in an expressive output space while maintaining modularity and tractability of training. The focus of this talk will be on some of the computational issues involved in these processes and those that arise from the need to do learning and inference with very large features and deal with very large amount of unstructured data.