Events Calendar

NL Seminar- Efficient Computation of Substring Posteriors from Lattices using Weighted Factor Automata

Speaker: Dogan Can, USC/SAIL

Talk Title: Efficient Computation of Substring Posteriors from Lattices using Weighted Factor Automata

Series: Natural Language Seminar

Abstract: Efficient computation of substring posteriors from lattices has applications in the estimation of document frequencies in spoken corpora and lattice-based minimum Bayes-risk decoding in statistical machine translation. In this talk, we present a new algorithm for exact substring posterior computation that leverages the following observations to speed up computation: i) the set of substrings for which the posteriors will be computed typically comprises all n-grams in the lattice up to a certain length, ii) posterior probability is equivalent to expected count for substrings that do not repeat on any path of the input lattice, iii) there are efficient algorithms for computing expected counts from lattices. We present experimental results comparing our algorithm with the best known algorithm in literature as well as a baseline algorithm based on finite state automata operations.



Biography: Dogan Can is a fifth year Ph.D. student at USC SAIL (Signal Analysis and Interpretation Lab). He works with Professor Shrikanth Narayanan on a range of topics including lattice indexing for spoken information retrieval, concurrent/online speech processing architectures and statistical modeling of psychotherapy sessions. His research interests include weighted finite state automata, automatic speech recognition, information retrieval, dialogue modeling and behavioral informatics.

Host: Nima Pourdamghani and Kevin Knight

More Info: http://nlg.isi.edu/nl-seminar/

Location: Information Science Institute (ISI) - 6th Flr Conf Rm # 689, Marina Del Rey

Audiences: Everyone Is Invited

Contact: Peter Zamar

Event Link: http://nlg.isi.edu/nl-seminar/